CADTH Reimbursement Review

Mogamulizumab (Poteligeo)

Sponsor: Kyowa Kirin Canada, Inc.

Therapeutic area: Mycosis fungoides, Sézary syndrome

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Introduction

Mycosis fungoides (MF) and Sézary syndrome (SS) are the 2 most common types of cutaneous T-cell lymphoma (CTCL), accounting for approximately one-half to two-thirds of CTCL cases (SS accounts for only 2% to 3% of all CTCLs).^1-5 The incidence of MF increases with age and is approximately 5.6 cases per 1 million people.^4-8 Clinically, SS affects mostly patients older than 60 years of age and is associated with a poor prognosis.^7,9 From 1992 through 2010 in Canada, there were 2,510 documented cases of MF (representing 37.5% of CTCL cases), and 110 cases of SS (representing 1.6% of CTCL cases).¹⁰

Both MF and SS result from infiltration of malignant T-lymphocytes, preferentially to skin. MF is usually associated with an indolent clinical course and intermittent, stable, or slow progression, while SS is a rare leukemic subtype of CTCL characterized by a more aggressive course of disease and shorter survival.¹¹ MF may be asymptomatic or manifest with symptoms of significant itch, pain, or burning of the skin. The clinical presentation of MF is highly variable, ranging from non-specific erythematous scaly patches to thin plaques or papulonodular eruptions, appearing mostly on non–sun-exposed areas.¹² In approximately 30% of cases, the patches or plaques evolve into tumours.¹³ Patients with SS present with a triad of skin redness (erythroderma), enlarged lymph nodes (lymphadenopathy), and a large number of circulating atypical lymphocytes (Sézary cells) in the skin, lymph nodes, and peripheral blood.^12,14 The psychological aspects of this incurable cancer affecting the skin present a significant burden to patients.¹⁵

As MF and SS can resemble other inflammatory dermatoses, diagnosis can be challenging, sometimes taking upward of 2 to 6 years.^16-21 Diagnosis of MF and SS is confirmed by skin biopsy and blood tests, though characteristic histologic features of MF may be absent in early disease.²² Secondary infections are a common complication related to an impaired skin barrier^23,24 and a dysregulated immune system due to the lymphoma itself and some immunosuppressive treatments.

Both MF and SS are incurable diseases; however, if discovered early, the prognosis for patients with early-stage MF can be very good.²⁵ Clinical staging of patients with MF is essential, not only for assessment of prognosis but also to guide decisions in the management of the disease. For patients with early-stage disease with limited skin involvement (stage IA or IB to IIA), skin-directed therapies are used, including topicals, phototherapy, and radiation therapy. Systemic therapy is recommended for advanced-stage disease (stage ≥ IIB). In advanced MF and SS, first-line treatment consists of oral retinoid therapy, with or without interferon, or single-agent chemotherapy, including methotrexate, gemcitabine, or liposomal doxorubicin. Second- and third-line treatment for MF consist of previously unused single-agent chemotherapy (anti-cancer) therapy, histone deacetylase (HDAC) inhibitors, or immunotherapies, such as brentuximab vedotin (in patients with CD30-positive disease). In SS, or for patients with blood involvement, extracorporeal photopheresis (ECP), with or without retinoids and interferon, may be used in either the first- or second-line setting; otherwise, eligible treatment options beyond the first-line setting remain the same as in MF.

Mogamulizumab has received a Health Canada Notice of Compliance for the treatment of adult patients with relapsed or refractory MF or SS after at least 1 prior systemic therapy. Mogamulizumab is a defucosylated, humanized immunoglobulin G1 (IgG1) kappa monoclonal antibody that binds to CC chemokine receptor 4 (CCR4), a G protein–coupled receptor for CC chemokines that is involved in the trafficking of lymphocytes to various organs and is expressed on the surface of some T-cell malignancies and on regulatory T cells and a subset of T helper 2 T cells.²⁶

The objective of the current review is to review the beneficial and harmful effects of mogamulizumab 1.0 mg/kg for the treatment of adults with relapsed or refractory MF or SS after at least 1 prior systemic therapy.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient and clinician groups that responded to CADTH’s call for patient input and from the clinical experts consulted by CADTH for the purpose of this review.

Patient Input

Between March 8, 2021, and September 21, 2021, Lymphoma Canada, in collaboration with the Canadian Skin Patient Alliance and Cutaneous Lymphoma Foundation, conducted an anonymous online survey of CTCL patients, primarily patients with MF or SS. A total of 449 patients who responded to the survey were diagnosed with either MF or SS; of these, 46 (14%) were living in Canada. The majority of respondents were aged 60 years or older (69%) and more than half were female (54%).

MF can be difficult to diagnose, with symptoms often occurring for many years before a diagnosis is made. It can masquerade as other more common skin conditions, such as eczema or psoriasis. The majority of patients participating in this survey were diagnosed between 1 and 5 years ago (41%), with a portion of patients diagnosed more than 10 years ago (26%). Only 23% of patients had their condition correctly diagnosed as MF or SS at presentation. Symptoms of MF or SS that most impacted patients’ quality of life (QoL) at diagnosis included visible patches or lesions (78%) and itchiness of skin or skin lesions (57%). Symptoms that most commonly affected respondents’ current QoL were similar to those at diagnosis, with the stress of diagnosis plus the fear of and worry about disease progression as well as anxiety impacting their current well-being.

A total of 327 patients provided information about their experience with CTCL treatments for MF and SS. Of the 211 patients who received systemic treatments, 27% received interferon, 26% received methotrexate, 24% received oral bexarotene, 23% received mogamulizumab, and 10% were treated with romidepsin. According to the patients who responded, the most common side effects they experienced during MF or SS CTCL treatments included fatigue (41%), skin pain or burning (34%), skin irritation or rash (33%), and moderate or severe itching (30%). A total of 27% of survey respondents indicated that the number of clinic visits had the most significant impact on their QoL, while 21% indicated that treatment-related fatigue had the most significant impact on their QoL, and cited additional negative impacts due to treatment on their ability to work (21%), travel (21%), and have intimate relationships (19%).

Access to treatment within the patient’s community is an important consideration, as certain treatments may be available only at specific tertiary cancer centres. A subanalysis of patients living in Canada revealed that 32% of them could not access treatment locally. Most patients responding to this survey were living in the US, where more treatment options are approved for use, including mogamulizumab; however, 81% of all respondents cited the importance of having an increased number of treatment options available. Patients indicated that longer survival (82%) and better QoL (76%) are extremely important outcomes for new therapies, followed by longer treatment-free periods (70%), easier treatment application (68%), and fewer side effects (64%).

Of all patients who responded to the survey, 52 (12%) had experience with mogamulizumab, accessed largely through private insurance (42%), public drug programs (29%), compassionate access programs (17%), or clinical trials (8%). At the time of the survey, 44% were still receiving treatment, 12% of patients had “completed their full course of mogamulizumab” [sic], 12% had discontinued due to side effects, and 10% did not have their symptoms controlled with mogamulizumab. Only 2 patients had their disease progress during treatment, while 25% were in remission. Of the 52 respondents, a total of 36% of patients had all of their symptoms managed by mogamulizumab, with major symptom management in skin itchiness (62%), red skin patches or rash (56%), and skin pain (25%). When asked to describe their experience with mogamulizumab, 69% of patients indicated they had a good to excellent experience with the therapy.

Clinician Input

Input From the Clinical Experts Consulted by CADTH

According to advice obtained from a clinical expert consulted by CADTH for this review, treatment selection in Canada is often guided by which drugs are funded and available rather than the one with the best chances of response and fewest side effects. They noted that most currently available treatment options demonstrate suboptimal response rates (< 30% to < 50%) or responses of limited duration (< 4 months), and acknowledged that these responses are primarily in the context of skin; however, when used in SS, they are in the context of blood involvement, as well. Per the clinical experts, in Canada, systemic therapy is generally limited to interferon, isotretinoin (off-label use for MF), or oral methotrexate. Bexarotene was noted by the experts as a retinoid that is superior to isotretinoin; however, it is not approved by Health Canada. Other options include local radiotherapy, ECP, chemotherapy (gemcitabine or liposomal doxorubicin), vorinostat, romidepsin, and pralatrexate, though many of these are not funded in Canada. Brentuximab was recently approved for use in CD30-expressing MF, limited or advanced stage, that has relapsed following one of these systemic approaches. The experts stated that some patients derive benefit from existing therapies before ultimately progressing and requiring an alternative, thus agreed that other systemic treatments, including interferon or retinoids with or without ECP (as available), should be attempted before mogamulizumab, based on availability.

The experts noted that diagnosis often takes upward of 3 years, with misdiagnosis occurring frequently early on. There are no pre-symptomatic tests to identify patients best suited for treatment with mogamulizumab. According to the experts, the patients most suitable for mogamulizumab are those with advanced-stage MF (stage IIIB, IVA, IVB) with or without blood involvement, or SS that has failed front-line systemic therapy. Conversely, the clinical experts indicated that patients with stage IA MF should not be treated with mogamulizumab, as they were not included in the pivotal clinical trial, and patients with stage IB, IIA, or IIB are less likely to derive benefit (as evidenced by the response outcomes demonstrated in the study), possibly due to the mechanism of action of mogamulizumab. Overall, the clinical experts felt that aside from patients with stage IA, who were not eligible for the MAVORIC trial, funding criteria should not include staging information, as the primary outcome of the MAVORIC trial demonstrated improved progression-free survival (PFS) across all included stages (IB to IV), but the study was not powered to detect differences by stage.

The experts expressed that the goals of treatment for MF and SS consist of prolonged survival, improved response rate, improvement in skin-related symptoms, and QoL. They noted that these outcomes are reflective of what is measured in clinical trials, as assessed using existing defined response criteria for MF and SS in any of each disease compartment (skin, lymph nodes, and blood). These also include criteria to define progressive disease (PD), and they noted that intolerable AEs, notably mogamulizumab-associated rash that does not respond to management algorithms or other AEs, would be reason to discontinue.

Clinician Group Input

The clinician group input was provided by 2 clinician groups: a network of Canadian cutaneous lymphoma providers and the Ontario Health (Cancer Care Ontario) Hematology Cancer Drug Advisory Committee. The clinician groups noted the individualized approach to treatment as well as the lack of a defined standard of care in MF or SS. Along with suboptimal response rates and limited duration with current treatments, access issues were cited as a major unmet need in this population. The clinician groups noted the lack of publicly funded treatments in Canada and that treatment selection is guided more by which drugs are funded and available rather than necessarily the one with the best chances of response and fewest side effects. The clinician groups expressed that patients should be offered mogamulizumab if they meet the eligibility criteria from the clinical trial, with 1 clinician group highlighting its use in advanced-stage MF, with or without blood involvement, and Sézary syndrome, and that mogamulizumab would likely be used in the second line following chemotherapy, brentuximab vedotin (if CD30-positive), or interferon. One clinician group considered the potential for mogamulizumab’s concurrent use with ECP, particularly for patients with SS. Important goals of treatment, as well as factors considered clinically meaningful to treatment outcomes, include disease control via a reduction in the frequency or severity of symptoms, improvement in symptoms, or stabilization of disease, prolonged survival, and improvement in QoL. Lastly, the clinician groups noted that mogamulizumab would be received in an outpatient setting, and highlighted that disease progression, AEs, and treatment-related toxicity should be considered when deciding to discontinue treatment.

Drug Program Input

The Provincial Advisory Group identified the following jurisdictional implementation issues: relevant comparators, considerations for initiation of therapy, considerations for prescribing of therapy, generalizability, funding algorithm, care provision issues, and system and economic issues. The clinical experts consulted by CADTH weighed evidence from the MAVORIC trial and used their clinical expertise to provide responses to the Provincial Advisory Group’s implementation questions. Refer to Table 5 for more details.

Clinical Evidence

Pivotal Studies and Protocol Selected Studies

Description of Studies

One study (MAVORIC) was included in the review. MAVORIC is an open-label, multi-centre, 1-way crossover, phase III randomized controlled trial (RCT) evaluating the effectiveness of mogamulizumab compared with vorinostat in patients with CTCL whose disease failed to respond to at least 1 prior course of systemic therapy. Patients in the MAVORIC trial were required to have a histologically confirmed diagnosis of MF or SS; stage IB, IIA, IIB, III, or IV disease; and an Eastern Cooperative Oncology Group (ECOG) Performance Status of 0 or 1. A total of 372 patients were randomized 1:1 to either mogamulizumab 1.0 mg/kg IV infusion on days 1, 8, 15, and 22 of the first cycle, and on days 1 and 15 of subsequent cycles (n = 186), or vorinostat 400 mg by mouth once daily on day 1 of each 28-day cycle (n = 186). The primary outcome of the MAVORIC trial was PFS, with key secondary end points of overall response rate (ORR) and health-related quality of life (HRQoL) as assessed by the Skindex-29, Functional Assessment of Cancer Therapy–General (FACT-G), and the 3-level EQ-5D questionnaire (EQ-5D-3L). Other secondary outcomes included best overall response (BOR) and duration of response (DOR). Overall survival (OS) was an exploratory outcome of the MAVORIC trial. MAVORIC was conducted in 11 countries; however, no Canadian investigative sites were included. The clinical data cut-off for the final analysis of the MAVORIC study was December 31, 2016. No interim analyses were conducted.

Demographic and baseline characteristics of the MAVORIC trial were well balanced: the median age was 64 years, 58.1% were male, and 69.9% were White. Most patients had an ECOG Performance Status of 0 (56.5%). Most patients had MF (54.8%), with the remaining patients diagnosed with SS (45.2%). The population in the MAVORIC trial was heavily pre-treated, with a median of 3 prior systemic therapies in each treatment arm.

Efficacy Results

ORR was higher in the mogamulizumab group compared with the vorinostat group; 52 patients (28.0%) treated with mogamulizumab achieved an ORR compared with 9 (4.8%) with vorinostat.²⁷ The results of subgroup and post hoc analyses also suggested that mogamulizumab provided superior response rates in patients with SS (37.0% versus 2.3%),²⁷ advanced-stage disease (30.0% versus 2.9%),²⁸ and B₂ blood involvement (37.4% versus 3.2%).²⁹ Additionally, 83 patients (66.9%) treated with mogamulizumab versus 23 patients (18.4%) treated with vorinostat achieved an ORR in the blood compartment, and 78 (41.9%) versus 29 (15.6%) patients achieved an ORR in the skin compartment.²⁷

At the time of the data cut-off (December 31, 2016), the median efficacy follow-up was 17.0 months. The median PFS was 7.70 months (95% confidence interval [CI], 5.67 to 10.33) in the mogamulizumab group, compared with 3.10 months (95% CI, 2.87 to 4.07) in the vorinostat group (P < 0.0001). The HR for mogamulizumab versus vorinostat was 0.53 (95% CI, 0.41 to 0.69), in favour of mogamulizumab. Results of sensitivity analyses varying the definition of PFS were consistent with the primary analysis.²⁷ Multiple subgroup and post hoc analyses of PFS were conducted, demonstrating improved survival with mogamulizumab in patients with SS (13.30 versus 3.13 months), advanced-stage disease (9.40 versus 3.07 months),²⁸ stage III or IV disease (10.90 versus 3.00 months),²⁷ and B₁ (8.63 versus 2.53 months) and B₂ (11.17 versus 3.30 months) blood involvement.²⁹

HRQoL was assessed using the Skindex-29, FACT-G, and EQ-5D-3L. In general, the magnitude of improvement from baseline to cycle 5 was greater with mogamulizumab compared with vorinostat. Across all scale domains for the Skindex-29, mogamulizumab and vorinostat reduced scores from baseline to cycle 5, with mean symptom scale scores demonstrating the greatest improvement (−18.0 versus −8.2). Results were consistent across HRQoL measures, with both mogamulizumab and vorinostat demonstrating an increase in total FACT-G score at cycle 5. With the EQ-5D Visual Analogue Scale (VAS), mean baseline scores increased from 60.9 (SD = 22.10) to 69.0 (SD = 20.30) at cycle 5 for mogamulizumab compared with 60.8 (SD = 20.02) to 63.5 (SD = 20.08) for vorinostat.^27,30

Other secondary outcomes evaluated also favoured mogamulizumab, with a BOR of 34.9% versus 6.5%, and a median DOR of 14.07 months with mogamulizumab compared with 9.13 months with vorinostat. At a median efficacy follow-up of 17.0 months, there was no difference in median OS between mogamulizumab (median OS was not estimable [NE]) and vorinostat (43.93 months).²⁷

Harms Results

The overall incidence of treatment-emergent adverse events (TEAEs) was consistent between the mogamulizumab (97.3%) and vorinostat (99.5%) arms. AEs were reported for randomized treatment. The incidence of specific TEAEs in patients who crossed over was not reported. The most frequently reported TEAEs with mogamulizumab were infusion-related reaction (33.2%), drug eruption (23.9%), diarrhea (23.4%), and fatigue (23.4%). The most frequently reported TEAEs with vorinostat were diarrhea (61.8%), nausea (42.5%), fatigue (37.6%), and thrombocytopenia (30.6%). The incidence of serious adverse events (SAEs) was higher in the mogamulizumab group compared with vorinostat (37.5% versus 24.7%). The most frequently occurring SAEs by system organ class (SOC) were infections and infestations (16.3% versus 10.8%). Withdrawals due to AEs were similar between treatment groups during the randomized treatment period, with 19.0% of patients in the mogamulizumab group and 23.1% of patients in the vorinostat group discontinued treatment due to AEs, most frequently due to drug eruption for mogamulizumab (7.1%). A total of 5 (2.7%) and 9 (4.8%) patients, respectively, died due to AEs during the randomized treatment period.²⁷

Notable harms, including infusion-related reactions (IRRs) and infections, were generally more frequent in the mogamulizumab group. In the randomized treatment phase, IRRs with mogamulizumab occurred in 33.2% of patients compared with 0.5% of patients receiving vorinostat, while infections and infestations occurred in 64.1% and 50% of patients, respectively. Though not expressly reported as a group in the MAVORIC trial, the most frequently occurring immune-related TEAEs with mogamulizumab were drug eruption (23.9% versus 0.5%) and fatigue (23.4% versus 37.6%).²⁷

Critical Appraisal

The MAVORIC trial was an open-label, phase III RCT. In general, patients did not differ with regard to baseline characteristics, indicating that randomization was successful. There were no notable differences between mogamulizumab and vorinostat among patients discontinuing randomized treatment, with most patients in both groups discontinuing due to disease progression. The reviewers and the clinical experts consulted by CADTH agreed the open-label design used was appropriate. However, they noted this could potentially increase the risk of bias in the reporting of outcomes that are subjective in measurement and interpretation, such as response, HRQoL, and AEs; however, a blinded independent review was instituted to assess for any potential investigator bias. The MAVORIC trial also included a 1-way crossover design, where patients with disease that failed treatment with vorinostat were able to cross over to the mogamulizumab group, which may have confounded the results for OS. As an exploratory end point, results for OS are considered descriptive The end points considered in the study were clinically appropriate for this population per the clinical experts consulted by CADTH and have been recommended by the International Society for Cutaneous Lymphomas (ISCL), the United States Cutaneous Lymphoma Consortium, and the Cutaneous Lymphoma Task Force of the European Organisation for Research and Treatment of Cancer (EORTC).³¹ Acceptable methods to account for multiplicity were used in the MAVORIC trial to adjust for the overall studywise type I error rate for the key secondary end points. All subgroup analyses of the MAVORIC trial, including disease type, stage, blood involvement, region, age, gender, race, and lactate dehydrogenase levels, were pre-specified; however, they were not controlled for multiplicity.

In discussion with the clinical experts consulted by CADTH, the inclusion and exclusion criteria for the MAVORIC trial were considered generally appropriate for enrolling patients with a rare disease. There were, however, no Canadian sites in the MAVORIC trial. The MAVORIC trial enrolled patients with MF or SS with stage IB, IIA, IIB, III, or IV disease. The inclusion of so many stages of disease was considered by the CADTH review team and clinical experts to be quite broad, given the differences in disease behaviour and prognosis at each stage. The clinical experts also noted that the trial included a high proportion of patients with SS compared with the overall incidence of SS in Canada. The clinical experts also expected more patients with an ECOG Performance Status of 2 in Canadian clinical practice, who were not eligible for the MAVORIC trial. Though approved by Health Canada for the treatment of CTCL, vorinostat is not widely accessible in Canada; thus, the comparative efficacy of mogamulizumab relative to vorinostat in patients in Canada with MF or SS may not be generalizable.

Results for HRQoL demonstrated improvements in HRQoL domains at various time points in the analysis; however, per the definitions of these populations, HRQoL results were presented only as observed data and not for the entire intention-to-treat (ITT) population. As such, these patients can be considered responders to treatment, which may bias the results. Moreover, high attrition rates were observed for all HRQoL measures, ranging from 42.2% to 44.8% of mogamulizumab-treated patients and 66.3% to 67.03% of vorinostat-treated patients failing to complete the assessments at 6 months compared with baseline. This created uncertainty in the results and thus may impact the generalizability of the results.

Indirect Comparisons

Description of Studies

The sponsor submitted an indirect treatment comparison (ITC) that compared individual patient data from the MAVORIC trial with the populations of relevant trial reports for comparators of interest in the treatment of CTCL, including the subtypes MF and SS. The objective of the sponsor-submitted report was to assess the feasibility of performing ITCs to compare mogamulizumab with relevant comparators of interest for the outcomes of PFS, OS, time-to-treatment failure (TTF), ORR, and Skindex-29 score. Of specific focus was the comparison between mogamulizumab (MAVORIC trial) and the drugs in the 2 arms of the ALCANZA study (brentuximab vedotin and physician’s choice [methotrexate or bexarotene]), given that these comparators were expected to be the main comparators in the cost-effectiveness model.

An initial feasibility assessment was conducted based on the findings of a clinical systematic literature review. The review extracted evidence from 39 publications detailing 26 different studies: 14 parallel trials and 12 single-arm studies. The feasibility assessment was done to determine whether a network meta-analysis was possible by evaluating the network connectivity of included trials via comparators, the trial inclusion and exclusion criteria (i.e., population), the demographic and disease characteristics, and the study end points.

Based on the comparators of interest considered in the feasibility assessment, no connected networks could be formed with mogamulizumab. As such, an unanchored matching-adjusted indirect comparison (MAIC) was used as an alternative analytical method to compare treatment from the ALCANZA trial with mogamulizumab. Multiple MAICs were conducted, including the comparison of the mogamulizumab arm with the brentuximab vedotin arm for the entire (ITT) population of the MAVORIC and ALCANZA trials, as well as the MF-only population of the MAVORIC trial. Additional comparisons included the vorinostat arm of the MAVORIC trial with the physician’s choice (methotrexate or bexarotene) arm in the ITT and MF-only populations. The outcomes evaluated included PFS, OS, and response in skin.

Efficacy Results

Two studies were included in the sponsor-submitted MAIC: MAVORIC and ALCANZA. For PFS, the results of the MAIC comparing mogamulizumab with brentuximab after weighting demonstrated a greater probability of PFS events with mogamulizumab over brentuximab using both the ITT (HR = 2.21; 95% CI, 1.68 to 3.19) and MF-only (HR = 2.52; 95% CI, 1.78 to 3.75) populations, while there was no difference between vorinostat and physician’s choice. For OS, there was no difference between mogamulizumab and brentuximab after matching patients to the ALCANZA population in the MAVORIC ITT (HR = 0.90; 95% CI, 0.62 to 1.27) or MF-only populations (HR = 0.79; 95% CI, 0.45 to 1.18). The comparison between vorinostat and physician’s choice could not be conducted due to the crossover in the trials. | | | |||||| | | ||| | ||||||||||| | |||||||||| | ||| | |||||| | | || | || | ||||||||| | ||||| | | || | | | | | | | || | || | | | ||| | |||| | | | | || | ||||||||| | |||||| | || | | | ||| | ||||| | ||| | ||| | | | | |||||||| | ||||| | |||||||| | | | ||||||||| |||||||

Critical Appraisal

The choice to conduct an MAIC between MAVORIC and ALCANZA was justified, considering the lack of a common comparator. Moreover, no rationale or justification for using brentuximab and physician’s choice from the ALCANZA trial as the primary comparator in the MAIC was provided. The main comparator for the MAIC, brentuximab vedotin, is indicated for adult patients with primary cutaneous anaplastic large-cell lymphoma or CD30-expressing MF who have received prior systemic therapy. This population was different from that of the MAVORIC trial, as no SS patients were included, and CD30 positivity was not assessed in MAVORIC, increasing the uncertainty in the analyses. The MAIC also provided a naive comparison between vorinostat in the MAVORIC trial and physician’s choice (consisting of methotrexate or bexarotene) in the ALCANZA trial. These 3 drugs (vorinostat, methotrexate, and bexarotene) are rarely used in Canadian clinical practice, according to the clinical experts consulted by CADTH, thus further limiting the generalizability of these results.

There were many key differences between the MAVORIC and ALCANZA trials that impact the comparability of populations within these studies, particularly the inclusion criteria of the studies (i.e., the eligible population), the specific diagnosis, as well as differences in various baseline characteristics, such as CD30 status, disease stage, blood involvement, prior treatments, and treatments in the study. The heterogeneity in population was accounted for by conducting matched analyses with the ITT population of MAVORIC as well as the MF-only population of MAVORIC to the ALCANZA population, resulting in an effective sample size that was reduced by 50.5% for the ITT population, and 25.7% for the MF-only population. There were also considerable differences in baseline age, ECOG Performance Status, disease stage, and blood involvement; however, it was uncertain as to what direction this may impact results. A comprehensive list of prognostic factors and treatment-effect modifiers for weighting was provided (including these baseline characteristics); however, the method of identification, justification, and validation of prognostic factors and treatment-effect modifiers was unclear, and it was uncertain whether all key factors were included in weighting; thus, the risk of bias on the relative treatment effects for unanchored MAICs is considered substantial and must be considered.

Overall, given the differences between MAVORIC and ALCANZA in terms of inclusion and exclusion criteria, the diagnosed population and the populations included in the analyses, the baseline characteristics, the differences in study design, and the large reduction in effective sample size, the results of the MAIC are uncertain and may not be generalizable.

Conclusions

Based on the clinical evidence from the phase III, open-label RCT, MAVORIC, conducted in patients with MF or SS whose condition has failed to respond to at least 1 prior systemic therapy, mogamulizumab was associated with a clinically meaningful and statistically significant improvement in PFS and ORR compared with vorinostat; however, given the open-label design of the study, these results must be interpreted with caution due to the potential for reporting biases. Notably, these results were most pronounced in patients with advanced SS and blood involvement. Patients highlighted the need for accessible, alternative treatment options that improve survival, and HRQoL; however, due to high attrition rates, the short follow-up duration of the MAVORIC trial, and the high rate of crossover, there remains uncertainty in the effect of mogamulizumab on HRQoL and OS, as no difference between mogamulizumab and vorinostat was observed for OS. The sponsor-submitted MAIC was associated with significant clinical heterogeneity in the included studies, which resulted in multiple limitations and notable uncertainty in the comparative efficacy of mogamulizumab. Given the difficulty in accessing treatment for MF and SS, the lack of standard of care, and that vorinostat is not widely available in Canada, there is some uncertainty around the true significance of the comparative efficacy of mogamulizumab. Overall, the available evidence suggests that mogamulizumab provides an additional treatment option for patients with MF or SS whose disease has failed to respond to at least 1 systemic therapy, delaying disease progression, and providing a meaningful clinical response in the overall population of both MF and SS.

Introduction

Disease Background

CTCLs are a group of extra-nodal non-Hodgkin lymphomas in which cancerous lymphocytes affect the skin. MF and SS are the 2 most common types of CTCL. Both MF and SS result from infiltration of malignant T-lymphocytes preferentially to skin. MF is usually associated with an indolent clinical course and intermittent, stable, or slow progression, while SS is a rare leukemic subtype of CTCL characterized by a more aggressive course of disease and shorter survival.¹¹ Together, MF and SS account for approximately one-half to two-thirds of CTCL cases (SS accounting for only 2% to 3% of all CTCLs).^1-5

MF is a highly symptomatic disease, with variable clinical presentation. The clinical presentation of MF is highly variable, ranging from non-specific erythematous scaly patches to thin plaques or papulonodular eruptions appearing mostly on non–sun-exposed areas.¹² In approximately 30% of cases, the patches or plaques evolve into tumours.¹³ In some instances, MF can spread to the lymph nodes, blood, or other organs, such as the spleen, liver, or lungs, though mucosal involvement is rare.^9,12 Patients with SS present with a triad of skin redness (erythroderma), enlarged lymph nodes (lymphadenopathy), and a large number of circulating atypical lymphocytes (Sézary cells) in the skin, lymph nodes, and peripheral blood.^12,14 The psychological aspects of this incurable cancer affecting the skin present a significant burden to patients.¹⁵

According to the National Cancer Institute, the incidence of MF has remained stable since 1995 with an incidence of 5.6 per million persons.⁵ In Canada, 2,510 cases of MF were documented from 1992 through 2010 representing 37.5% of CTCL cases, and 110 cases of SS representing 1.6% of CTCL cases.¹⁰ The incidence of MF increases with age, with most patients diagnosed in their 50s and 60s, with an increased incidence in patients greater than 70 years.^4,6-8 MF is more common in Black patients than Caucasians, and males, with a male-to-female ratio of 1.6 to 2:1.^2,4,22 Due to the indolent course of the disease, the median survival in early-stage disease ranges from 10 to 35 years, dropping to approximately 4 years with advanced disease.^7,32,33 Clinically, SS mostly affects patients > 60 years of age and is associated with a poor prognosis (median survival of approximately 3 years).^7,9

As MF and SS can resemble other inflammatory dermatoses, diagnosis can be challenging, sometimes taking upwards of 2 to 6 years.^16-21 Diagnosis of MF or SS is confirmed by skin biopsy and blood tests, though characteristic histologic features of MF may be absent in early disease.²² Diagnosis is further confirmed with immunophenotyping and DNA analysis of the T-cell receptor gene rearrangement to define the clonal population (CD2, CD3, CD4, CD5, CD7, CD8, CD20, and CD30).^5,12 Patients with MF and SS are typically characterized by the following immunophenotype: positive for CD2, CD3, CD5, and CD4; negative for CD8; and lack of certain T-cell markers (e.g., negative for CD7 and CD26 expression).^12,34

MF and SS are classified using a CTCL-specific modification of the tumour, nodes, metastasis, and blood classification system.^4,16 Both MF and SS are defined histologically and staged by the same criteria.^2,4 Patients are classified based on the number and type of skin lesions they have (T), lymph node involvement (N), metastasis or visceral organ involvement (M), and peripheral blood involvement (B), resulting in a diagnosis of a disease stage from IA through to IVB (Table 3).⁴ These stages can be grouped as early (stages IA, IB, and IIA) and advanced (stages IIB to IVB) disease.^4,18 Approximately two-thirds of MF patients have early-stage disease at the time of diagnosis.^18-20

Standards of Therapy

Both MF and SS are incurable diseases though, if discovered early, the prognosis for patients with early-stage MF can be very good.²⁵ Most treatment options do not result in durable remissions and are often given in an ongoing or maintenance fashion to achieve disease control with as little impact on QoL as possible. Clinical staging of patients with MF is essential, not only for assessment of prognosis but also to guide decisions in management of the disease. Selection of systemic therapies with lower toxicity treatments versus other systemic regimens is dependent on clinical features such as the extent of patches or plaques, disease burden profile in the skin, lymph nodes, and blood; prior therapies; comorbidities; and pathologic features, including confirmation of large-cell transformation (LCT) or folliculotropic MF and immunohistochemistry data (i.e., CD30-positive). A disease that relapses (at the same stage) after discontinuation of therapy often responds well to re-treatment with previous therapy.²⁵

For patients with early-stage disease with limited skin involvement (stage IA or IB to IIA), skin-directed therapies including topicals (topical nitrogen mustard, compounded carmustine, or imiquimod), phototherapy (psoralen plus UVA [PUVA] or narrow-band UVB light) or radiation, including external beam radiation therapy such as involved-site radiation therapy or total skin electron beam therapy (TSEBT) are used, which may provide disease control without major cumulative toxicities; however, its use is limited by the lack of availability of this technique in Canada. In cases of resistant or progressive skin disease, biologic response modifiers such as retinoids (isotretinoin, alitretinoin, bexarotene) or interferon may be combined with topical therapies.¹⁶

Systemic therapy is recommended for advanced-stage disease (stage ≥ IIB). In advanced MF and SS, first-line treatment consists of oral retinoid therapy (e.g., bexarotene, isotretinoin) with or without interferon, or single-drug chemotherapy or other anti-cancer therapies, which are generally preferred to combination chemotherapy (i.e., cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP] or cyclophosphamide, doxorubicin, vincristine, etoposide, and prednisone [CHEOP]). Single-drug chemotherapy for MF and SS includes methotrexate, gemcitabine, or liposomal doxorubicin, with pralatrexate or chlorambucil used in some cases. Second- and third-line treatment for MF consist of previously unused single-drug chemotherapy or other anti-cancer therapy, HDAC inhibitors such as vorinostat or romidepsin, or immunotherapies such as brentuximab vedotin (in patients with CD30-positive disease), or alemtuzumab. In SS, or in patients with blood involvement, ECP with or without retinoids and interferon may be used in either the first- or second-line setting; otherwise, eligible treatment options beyond the first-line setting remain the same as in MF. Multi-drug chemotherapy regimens are generally reserved for disease refractory to multiple prior therapies or for bulky lymph node or solid organ disease, and/or as a bridge to allogeneic hematopoietic stem-cell transplant, given the greater toxicity.¹⁶ In many cases, given the lack of reliable options, enrolment in a clinical trial may be pursued.^16,25 Systemic therapy may often be combined with skin-directed therapy to maximize clinical responses in the skin compartment and to provide added efficacy without cumulative toxicities.²⁵

Drug

Mogamulizumab is a defucosylated, humanized IgG1 kappa monoclonal antibody that binds to CCR4, a G protein–coupled receptor for CC chemokines that is involved in the trafficking of lymphocytes to various organs. Non-clinical in vitro studies demonstrate that mogamulizumab binding targets a cell for antibody-dependent cellular cytotoxicity, resulting in depletion of the target cells. CCR4 is expressed on the surface of some T-cell malignancies and is expressed on regulatory T cells and a subset of T helper 2 T cells.²⁶

Mogamulizumab is provided as a 20 mg/5 mL vial for reconstitution at a dose of 1.0 mg/kg. Mogamulizumab is administered as an IV infusion over at least 60 minutes at a dose of 1 mg/kg on days 1, 8, 15, and 22 of the first 28-day cycle, then on days 1 and 15 of each subsequent 28-day cycle until disease progression or unacceptable toxicity.²⁶

Mogamulizumab has received a Notice of Compliance (NOC) from Health Canada for the treatment of adult patients with relapsed or refractory MF or SS after at least 1 prior systemic therapy.²⁶ The Health Canada NOC was granted on June 2, 2022.³⁵ Mogamulizumab has not been previously reviewed by CADTH. The sponsor’s reimbursement request is the same as the proposed Health Canada indication.

Mogamulizumab was approved by the Ministry of Health, Labour, and Welfare in Japan for the treatment of primary and relapsed or refractory, CCR4-positive adult T cell leukemia/lymphoma (December 14, 2014, and March 30, 2012, respectively), relapsed or refractory, CCR4-positive peripheral T cell lymphoma (March 17, 2014); and relapsed or refractory CCR4-positive CTCL (March 17, 2014). Mogamulizumab has been granted an orphan designation for CTCL in the EU, Japan, and the US.²⁷

Key characteristics of mogamulizumab and the main comparators as specified in the protocol for this review (Table 6) are summarized in Table 4.

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

Lymphoma Canada, in collaboration with the Canadian Skin Patient Alliance and Cutaneous Lymphoma Foundation, conducted an anonymous online survey between March 8, 2021, and September 21, 2021. The respondents consisted of patients with CTCL, primarily MF or SS. A total of 449 patients who responded to the survey had been diagnosed with either MF or SS; 46 (14%) of these patients were living in Canada. The majority of respondents were aged 60 years or older (69%), and more than half were female (54%).

MF can be difficult to diagnose, with symptoms that can occur for many years before a diagnosis is made. It can masquerade as other more common skin conditions, such as eczema or psoriasis. The majority of patients participating in this survey were diagnosed 1 to 5 years ago (41%), with a portion of patients diagnosed more than 10 years ago (26%). Only 23% of patients had their condition correctly diagnosed as MF or SS at presentation. Symptoms of MF and SS that most impacted patients’ QoL at diagnosis included visual patches or lesions (78%), and itchiness of skin or skin lesions (57%). Symptoms that most commonly affected respondents’ current QoL were similar to those at diagnosis, with the stress of diagnosis plus the fear of and worry about disease progression as well as anxiety impacting their current well-being.

A total of 327 patients provided information about their experience with CTCL treatment for MF and SS. Of the 211 patients who received systemic treatments, 27% received interferon, 26% received methotrexate, 24% received oral bexarotene, 23% received mogamulizumab, and 10% were treated with romidepsin. The most common side effects respondents experienced by patients during their MF or SS CTCL treatments included fatigue (41%), skin pain or burning (34%), skin irritation or rash (33%), and moderate or severe itching (30%). A total of 27% of survey respondents indicated that the number of clinic visits had the most significant impact on their QoL, while 21% indicated that treatment-related fatigue had the most significant impact on their QoL, citing additional negative impacts due to treatment on their ability to work (21%), travel (21%), and have intimate relationships (19%).

Access to treatment within the patient’s community is an important consideration, as certain treatments may be available only at specific tertiary cancer centres. A subanalysis of patients living in Canada revealed that 32% of them could not access treatment locally. Most patients responding to this survey were living in the US, where more treatment options are approved for use, including mogamulizumab; however, 81% of all respondents cited the importance of having an increased number of treatment options available. Patients indicated that longer survival (82%) and better QoL (76%) are extremely important outcomes for new therapies, followed by longer treatment-free periods (70%), easier treatment application (68%), and fewer side effects (64%).

Of all patients who responded to the survey, 52 (12%) had experience with mogamulizumab, accessed largely through private insurance (42%), public drug programs (29%), compassionate access programs (17%), or clinical trials (8%). At the time of the survey, 44% were still receiving treatment, 12% of patients “completed their full course of mogamulizumab” [sic], and 12% discontinued due to side effects, and 10% did not have their symptoms controlled with mogamulizumab. Only 2 patients had their disease progress during treatment, while 25% were in remission. Of the 52 respondents, a total of 36% of patients had all of their symptoms managed by mogamulizumab, with major symptom management in skin itchiness (62%), red skin patches or rash (56%), and skin pain (25%). When asked to describe their experience with mogamulizumab, 69% of patients indicated they had a good to excellent experience with the therapy.

Clinician Input

Input From the Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of MF and SS.

Unmet Needs

The clinical experts highlighted that MF or SS is an incurable disease that is often characterized by extensive and disfiguring cutaneous involvement with intense skin-related symptoms that include itching, burning, and pain. They emphasized that treatment selection in Canada is often guided by which drugs are funded and available rather than the 1 with the best chances of response and fewest side effects. The clinical experts stated that many of the treatments (TSEBT, ECP, or UV light) must be delivered at specialized or urban centres and thus are not practically accessible to many patients living with cutaneous lymphoma.

Per the clinical experts, the treatment goals for MF and SS consist of prolonged survival, improved response rates, improvement in skin-related symptoms, and QoL. It was highlighted by the experts that many currently available treatment options have suboptimal response rates (< 30% to < 50%), or responses of limited duration (< 4 months), and they acknowledged that these responses are primarily in the context of skin; however, when used in SS, they are in the context of blood involvement as well. They noted that some treatments may improve objective parameters, but patients may continue to experience skin symptoms or impairment of QoL or function. Additional goals of treatment noted by the experts included improved independence and reducing the burden on caregivers.

Place in Therapy

According to the clinical experts consulted by CADTH, limited-stage MF is treated with skin-directed therapy, including topical corticosteroids, UV light therapy, which is either narrow-band UVB or PUVA, and less commonly topical carmustine or other topical drugs. In some patients who are refractory or intolerant to these approaches, systemic therapy may be required. In Canada, this generally is limited to interferon, isotretinoin (off-label use for MF), or oral methotrexate. Bexarotene was noted by the experts as a retinoid that is superior to isotretinoin; however, it is not approved by Health Canada and is not listed on provincial cancer formularies. Advanced-stage MF involving tumours may be treated with local radiotherapy and systemic therapies similar to those used for limited-stage disease. Advanced MF with extra-cutaneous nodal or visceral involvement requires systemic therapies such as chemotherapy (gemcitabine, or liposomal doxorubicin). The experts expressed that many novel drugs that are recommended by international guidelines, such as vorinostat, romidepsin, and pralatrexate, are not funded in Canada. Brentuximab was recently approved for use in CD30-expressing MF, limited or advanced stage, that has relapsed following 1 of these systemic approaches. In SS, a front-line approach includes twice-monthly ECP, which is only available in some specialized urban centres. Systemic therapies such as interferon and isotretinoin are typically added upfront, then further treatments are limited by available funded options. This may include romidepsin, alemtuzumab, or pembrolizumab.

The experts noted that mogamulizumab was not studied in the upfront setting and that it would be used, as in the clinical trial, in the second line and beyond in patients who were intolerant or whose condition was refractory to first-line systemic treatments. The experts stated that some patients derive benefit from existing therapies before ultimately progressing and requiring an alternative treatment; thus, they agreed that other systemic treatments should be attempted before mogamulizumab, based on availability. Currently, mogamulizumab is only available to patients via special access programs. The experts agreed it would likely cause a shift in current therapy and would be considered after chemotherapy, brentuximab (in patients with CD30-positive disease), or interferon for advanced MF, or following interferon and/or a retinoid (isotretinoin) for SS.

The clinical experts highlighted the mechanism of action of mogamulizumab, considering its modification of the underlying disease mechanism, and the potential impact of its concurrent use with ECP.

Patient Population

The approach for managing MF and SS is individualized based on the extent of disease or stage, the goals of therapy, access to treatment, patient tolerance, and practical considerations related to administration. The experts noted that diagnosis can be challenging, often taking approximately 3 years, with misdiagnosis occurring frequently early on. As such, the experts agreed there was no pre-symptomatic testing that would identify the patients best suited for treatment with mogamulizumab.

The experts highlighted that the group of patients most suitable for mogamulizumab are those with advanced-stage MF (stage IIIB, IVA, IVB) with or without blood involvement, or SS that has failed to respond to front-line systemic therapy. These patients are also at the highest need of intervention, as they typically have the greatest symptom burden and a worse prognosis. Overall, the clinical experts felt that aside from patients with stage IA, who were not eligible for the MAVORIC trial, funding criteria should not include staging information, as the primary outcome of the MAVORIC trial demonstrated improved PFS across all included stages (IB to IV) and was not powered to detect differences by stage.

Experts expressed that currently available treatment options have suboptimal response rates and duration. Also, these patients sometimes cannot access TSEBT or ECP, so systemic therapy would be a better option. They are a subpopulation of the broader group of patients with limited-stage MF, who predicably progress to advanced stages over many years or decades.

According to the clinical experts, the patients not suitable for treatment with mogamulizumab are those who are currently benefiting from first-line treatment and those who are scheduled to have an allogenic transplant within 3 months of receiving the drug. It was also noted by the clinical experts that patients with stage IA MF were not included in the pivotal clinical trial, so should not be treated with mogamulizumab, as these patients, and those with stage IB, IIA, or IIB (as evidenced by the response outcomes demonstrated on study), are less likely to derive benefit due to the mechanism of action of mogamulizumab.

Assessing Response to Treatment

In general, the outcomes assessed in clinical practice reflect those typically measured in clinical trials using the defined response criteria for MF and SS found in the literature.³¹ One expert noted that response in clinical practice should be assessed monthly or every 15 days initially. Clinically meaningful response is generally measured as compartmentalized and composite responses, with a reduction in the severity of symptoms and an increase in PFS considered most important. An objective improvement of skin typically corresponds with improvement of other measures of disease burden in blood, lymph nodes, and viscera. Skin examination is assessed by an individual with expertise, generally using the modified Severity-Weighted Assessment Tool (mSWAT) score (a validated assessment tool).³¹ Subjective improvement in QoL assessments (e.g., Skindex-29) can be considered a clinically meaningful end point, with improvement in specific symptom parameters such as pruritis (itch) considered very crucial for some patients living with MF or SS. Additional serial diagnostic imaging, in the case of nodal and other extra-cutaneous disease compartments, and complete blood count (with or without flow cytometry for the count of Sézary cells) were noted as tools by the experts to determine response.

Discontinuing Treatment

According to the clinical experts, major factors to be considered when discontinuing treatment include disease progression in any of the 4 disease compartments (skin, lymph nodes, viscera, and blood). The experts noted there are published criteria for PD, in which case, discontinuation would be advised. Additionally, the experts noted that intolerable AEs, notably mogamulizumab-associated rash that does not respond to management algorithms, or other unacceptable AEs would be a reason to discontinue. It was proposed by the experts that one may consider continuing therapy with stable disease if the treatment is tolerated.

Prescribing Conditions

The clinical experts stated that the diagnosis of CTCL (MF and SS) is complex and requires clinicopathological correlation by experts in the field. Patients are most likely identified and under the care of a multidisciplinary group of doctors that may include but is not limited to hematologists, medical oncologists, dermatologists, radiation oncologists, hematopathologists, and dermatopathologists. Experts believe that mogamulizumab would typically be administered in an outpatient, ambulatory setting at an oncology or infusion centre under the supervision of trained personnel.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups.

The clinician group input was provided by 2 clinician groups: a network of Canadian cutaneous lymphoma providers and the Ontario Health (Cancer Care Ontario) Hematology Cancer Drug Advisory Committee. Clinician groups noted the individualized approach to treatment as well as the lack of a defined standard of care for MF or SS. Along with suboptimal response rates and the limited duration with current treatments, access issues were cited as a major unmet need in this population. The clinician groups highlighted the lack of publicly funded treatments in Canada and noted that treatment selection is guided more by which drugs are funded and available rather than necessarily the one with the best chances of response and fewest side effects. The clinician groups stated that patients should be offered mogamulizumab if they meet the eligibility criteria from the clinical trial, with 1 clinician group highlighting its use in advanced-stage MF (with or without blood involvement) or SS, and that mogamulizumab would likely be used in the second line following chemotherapy, brentuximab vedotin (if CD30-positive), or interferon. One clinician group considered the potential for mogamulizumab’s concurrent use with ECP, particularly for patients with SS. Important goals of treatment, as well as outcomes of treatment considered clinically meaningful, include disease control via a reduction in the frequency or severity of symptoms, an improvement in symptoms or stabilization of disease, prolonged survival, and an improvement in QoL. Lastly, the clinician groups noted that mogamulizumab would be received in an outpatient setting and highlighted that disease progression, AEs, and treatment-related toxicity should be considered when deciding to discontinue treatment.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 5.

The Provincial Advisory Group’s implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 5.

Table 5: Summary of Drug Plan Input and Clinical Expert Response

CNS = central nervous system; ECOG PS = Eastern Cooperative Oncology Group Performance Status; HC = Health Canada; MF = mycosis fungoides; NYHA = New York Heart Association; pCPA = pan-Canadian Pharmaceutical Alliance; pERC = CADTH pan-Canadian Oncology Drug Review Expert Review Committee; PFS = progression-free survival; SS = Sézary syndrome.

Clinical Evidence

The clinical evidence included in the review of Poteligeo (mogamulizumab) is presented in 3 sections. The first section, the systematic review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review. The third section includes sponsor-submitted long-term extension studies and additional relevant studies that were considered to address important gaps in the evidence included in the systematic review.

Systematic Review (Pivotal and Protocol Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of mogamulizumab 1.0 mg/kg for the treatment of adults with MF or SS who have received at least 1 prior systemic therapy.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 6. Outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

Of note, the systematic review protocol presented in Table 6 was established before the granting of an NOC from Health Canada.

The literature search was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist.⁴¹

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946—) through Ovid and Embase (1974—) through Ovid. All Ovid searches were run simultaneously as a multi-file search. Duplicates were removed using Ovid deduplication for multi-file searches, followed by manual deduplication in Endnote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concept was mogamulizumab. Clinical trial registries were searched: the US National Institutes of Health’s clinicaltrials.gov, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit the retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. See Appendix 1 for the detailed search strategies.

The initial search was completed on October 12, 2021. Regular alerts updated the search until the meeting of the CADTH pan-Canadian Oncology Drug Review Expert Committee (pERC) on February 9, 2022.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature tool.⁴² Included in this search were the websites of regulatory agencies (US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. See Appendix 1 for more information on the grey literature search strategy.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

Findings From the Literature

A total of 376 studies were identified from the literature for inclusion in the systematic review (Figure 1). The included studies are summarized in Table 7. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Description of Studies

One study was included in the review. The MAVORIC study (Study 0761-010) is an open-label, multi-centre, phase III RCT evaluating the effectiveness of mogamulizumab compared with vorinostat in patients with CTCL whose disease has failed to respond to at least 1 prior course of systemic therapy. The primary objective of MAVORIC was to compare the PFS of mogamulizumab versus vorinostat for patients with relapsed or refractory CTCL.²⁷

Using the ClinTrak interactive voice or web response system (CTIVRS), a total of 372 patients were randomized 1:1 via screening numbers assigned by the investigator (or designee) to either mogamulizumab 1.0 mg/kg IV infusion on days 1, 8, 15, and 22 of the first cycle and on days 1 and 15 of subsequent cycles (n = 186), or vorinostat 400 mg by mouth once daily on day 1 of each 28-day cycle (n = 186). Randomization was stratified by disease type (MF or SS) and disease stage (IB or II versus stage III or IV). One-way crossover from vorinostat to mogamulizumab was permitted in the case of disease progression or intolerable toxicity. Screening assessments were performed within 30 days before the start of study treatment. The clinical data cut-off for the final analysis of the MAVORIC study was December 31, 2016. No interim analyses were conducted.²⁷ MAVORIC was conducted in 11 countries; however, no Canadian investigative sites were included.

Populations

Inclusion and Exclusion Criteria

The inclusion and exclusion criteria for the MAVORIC trial are summarized in Table 7. Patients were required to have a histologically confirmed diagnosis of MF or SS; stage IB, IIA, IIB, III, or IV disease that had failed to respond to at least 1 prior systemic therapy; and an ECOG Performance Status of 0 or 1.²⁷

Patients with stage IA were not included in the MAVORIC trial. Due to the potential increased risk of post-transplant complications (i.e., graft-versus-host disease), patients were excluded from entry into the MAVORIC trial if they had already received an allogenic hematopoietic stem-cell transplant. Patients who received a transplant after treatment with mogamulizumab are being followed for details and outcome for at least 180 days post transplant.²⁷

Baseline Characteristics

Baseline characteristics for the MAVORIC trial are summarized in Table 8. Demographic and baseline characteristics of the ITT population were well balanced between arms: the median patient age was 64 years, 58.1% of patients were male and 69.9% were White. There was a higher representation of patients who were Black or African American in the mogamulizumab group compared with the vorinostat group (12.9% versus 7.0%). The majority of patients in each group had an ECOG Performance Status of 0 (57.0% and 55.9% in the mogamulizumab and vorinostat arms, respectively).²⁷

In the mogamulizumab group, 105 (56.5%) and 81 (43.5%) patients had MF or SS, respectively, while 99 (53.2%) and 87 (46.8%) patients in the vorinostat group had MF or SS. The proportion of patients with various clinical stages was well balanced between arms. There were 140 patients (37.7%) with stage IB to IIB, while 232 (62.4%) had stage IIIA to IV. The majority of patients had IVA₁: 82 (44.1%) versus 73 (39.2%) in the mogamulizumab and vorinostat arms, respectively. All patients in both arms had skin involvement, while 124 (66.7%) and 122 (65.6%) in the mogamulizumab group had nodal and blood involvement at baseline compared with 122 patients (65.6%) for each category in the vorinostat group. Visceral involvement was rare, occurring in 6 patients overall.²⁷

The population in the MAVORIC trial in each treatment arm had received a median of 3 prior systemic therapies. In total, 68 patients (18.3%) received 1 prior systemic therapy, 155 (41.7%) had received 2 to 3 prior systemic therapies, and 148 (39.9%) had received 4 or more prior systemic therapies.²⁷

Table 8: Summary of Baseline Characteristics (ITT)

CCR4 = CC chemokine receptor 4; CTCL = cutaneous T-cell lymphoma; ECOG PS = Eastern Cooperative Oncology Group Performance Status; ITT = intention-to-treat population; LDH = lactate dehydrogenase; MF = mycosis fungoides; SD = standard deviation; SS = Sézary syndrome.

^aNot reported for those countries that do not allow race or ethnicity to be collected.

^bTwo patients had an ECOG PS of 1 at pre-treatment but an ECOG PS of 2 on cycle 1, day 1.

Source: MAVORIC (0761 to 010) Clinical Study Report.²⁷

Interventions

Mogamulizumab

Mogamulizumab was administered via IV infusion on an outpatient basis at a dose of 1.0 mg/kg over at least 1 hour on days 1, 8, 15, and 22 of the first cycle and on days 1 and 15 of subsequent cycles until disease progression according to a global composite scoring system or unacceptable toxicity. Each treatment cycle was 28 days. Patients were observed for 1 hour following administration.²⁷

The start of a new cycle could be delayed for up to 2 weeks after completion of the prior cycle to allow resolution of treatment-related toxicities or for non-medical reasons. A delay of the start of a new cycle of greater than 2 weeks was to be discussed with the medical monitor or designee. No dose modifications of mogamulizumab were permitted in this study.²⁷

Mogamulizumab was given under the direct supervision of study personnel. A weight-based volume of mogamulizumab was diluted into 250 mL of saline (0.9% sodium chloride) and then filtered through a 0.22 μm protein-sparing (low protein binding) in-line filter.²⁷

It was recommended that patients be premedicated with acetaminophen or paracetamol orally and diphenhydramine 50 mg IV (or equivalent antihistamine) before the first mogamulizumab infusion.²⁷

Vorinostat

Vorinostat was administered on an outpatient basis at a dose of 400 mg orally once daily with food (4 × 100 mg capsules) until disease progression or unacceptable toxicity. Each treatment cycle was 28 days. The prescribing information used in the US was followed for recommendations regarding dose modification of vorinostat, including dose omission, reduction, or interruption for management of severe or intolerable adverse reactions.²⁷

Patients randomized to the vorinostat arm who had received 2 full treatment cycles and demonstrated progression of disease at the 8-week (cycle 2, day 26 to 28) assessment, or anytime thereafter, were allowed to cross over to treatment with mogamulizumab after discussion with the medical monitor or designee and receipt of approval from the sponsor. All summaries for the crossover portion were purely descriptive. No statistical comparisons were made for data collected from patients who crossed over from vorinostat to mogamulizumab.²⁷

Patients were required to return pill bottles at each study visit. Any unused capsules were counted.²⁷

Prior and Concomitant Therapy

While on study treatment, patients were not permitted to receive any experimental therapy or anti-cancer therapy other than the study medication. Any patient requiring other anti-cancer therapy was to be removed from study treatment.²⁷

Permitted treatments during the study included:²⁷

• hematopoietic growth factors and erythropoiesis-stimulating drugs

• patients taking low-dose systemic corticosteroid (≤ 20 mg prednisone equivalent) for at least 4 weeks before the pre-treatment visit were allowed to continue use; patients were allowed to receive intra-articular corticosteroid injections; intraocular, inhaled, or nasal corticosteroids; and replacement doses of systemic corticosteroids, as needed

• patients on a stable dose of medium or low-potency topical corticosteroids for at least 4 weeks before the pre-treatment visit could continue use at the same dose, although the investigator was to attempt to taper the medication to the lowest dosage tolerable while on study

• prophylactic treatment for infectious complications of T-cell lymphoma was permitted at the discretion of the investigator

• patients receiving vorinostat and concomitant therapy with coumarin-derivative anticoagulants were to be monitored for prolongation of prothrombin time and international normalized ratio results

• influenza vaccination consisting of killed virus or viral particles was allowed to be administered to the patients in the study.

Initiation of treatment with systemic steroids or an increase in the dose of steroids while on study was not permitted except to treat an infusion reaction. Patients who required systemic steroid therapy to treat a severe skin rash were to be discontinued from the study. Initiation of treatment with topical corticosteroids while on study was not permitted, except to treat an acute rash.²⁷

Other prohibited treatments during the study included:²⁷

• other concurrent HDAC inhibitors including valproic acid

• immunomodulatory drugs such as methotrexate, azathioprine, IV immunoglobulin, cyclophosphamide, cyclosporine, mycophenolate, infliximab, etanercept, leflunomide, adalimumab, lenalidomide, abatacept, rituximab, anakinra, interferon alpha, interferon beta, interleukin-2, and natalizumab

• any experimental therapy or anti-cancer therapy, including radiation and phototherapy, other than the study medications

• any live or live attenuated vaccine

• alternative medicines, particularly St. John’s wort.

The prior CTCL medications used by patients at baseline are summarized in Table 9.

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 10. These end points are further summarized after the table. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 4.

Primary Efficacy Outcome

The primary outcome of the MAVORIC trial was investigator-assessed PFS, defined as the time from randomization until documented PD in any compartment per the global composite CTCL response criteria (Table 11 and Table 12) or death due to any cause. The data cut-off date for the MAVORIC study was December 31, 2016. The primary efficacy analysis of the MAVORIC trial was conducted on randomized patients and does not include patients who crossed over from vorinostat to receive mogamulizumab.²⁷

Patients who discontinued study treatment for reasons other than disease progression were contacted every 3 months (± 14 days) until documented disease progression or death, or initiation of alternative therapy. Patients who discontinued treatment after achieving a complete response (CR) were assessed every 8 weeks (± 14 days) for the first 6 months (if a CR was achieved before 1 year on study) and then every 16 weeks (± 14 days) thereafter until progression. All patients or their referring physician were contacted every 3 months (± 14 days) to ascertain OS status.²⁷

The following censoring rules were applied for PFS:²⁷

• In the event that a randomized patient withdrew from the study for any reason before documented progression, the time from the day of randomization to the last post-baseline tumour assessment from any compartment (skin, blood, bone marrow, lymph nodes, or viscera) was used as a censored time point.

• For patients who were randomized to a treatment arm but had an unknown baseline assessment for a compartment, the PFS time was censored at the randomization date if there was no post-baseline tumour assessment for that compartment or if there was any evidence of lymphoma in that compartment at the post-baseline evaluation.

• For patients randomized to a treatment arm who withdrew from treatment before the first post-baseline tumour assessment for any reason other than disease progression, the PFS time was censored at the last documented visit.

• For patients who initiated a new anti-cancer therapy (including crossover to mogamulizumab) in the absence of a PFS event, the PFS time was censored at the last tumour assessment (for any compartment) before the start of the new anti-cancer therapy.

• For patients not known to have died or have documented progression as of the data cut-off date for the primary analysis, the PFS time was censored at the date of the last tumour assessment (for any compartment) before data cut-off.

Secondary Efficacy Outcomes

The key secondary outcomes of the MAVORIC trial included investigator-assessed ORR (i.e., a confirmed CR or partial response [PR]) and change from baseline in HRQoL.²⁷

Clinical Response (Overall Response Rate)

Prior to the start of treatment, skin disease (lesions and erythema) and blood were evaluated by the investigator using the mSWAT, and flow cytometry. The mSWAT is an objective, quantitative, severity-weighted method to assess the extent of skin lesions. Assessment of disease in skin and blood was evaluated at the end of each cycle. Further details on the mSWAT are reported in Appendix 4. Assessment of disease in lymph nodes and viscera was conducted by CT scan) and was performed at the end of cycle 1 (4 weeks), and every 8 weeks thereafter during the first year of treatment, and every 16 weeks thereafter. ORR was defined as the proportion of patients who were responders (confirmed CR or PR) based on investigator’s assessment per a global composite response score (Table 11) that was based on a response in each compartment (skin, blood, lymph nodes, and viscera) that was subsequently confirmed by 2 or more consecutive observations for a minimum of 4 weeks. The global composite response was assessed at 4 weeks (end of cycle 1) and every 8 weeks thereafter (cycle 3, 5, and so forth). After the first year, global composite response was assessed every 16 weeks (cycle 17, 21, and so forth). Individual criteria for measurement of response in skin, as determined by changes in mSWAT score, blood, viscera, and lymph nodes are summarized in Table 12. All individual response outcomes were investigator-assessed. Node, viscera, and blood response contributed toward global response only if disease was present at baseline unless there was global PD, which was defined as new disease in a previously uninvolved compartment.²⁷

Patients who met the criteria for PD in any compartment, or had a relapse in the blood, lymph nodes, or viscera confirmed at 2 consecutive visits, were to be discontinued from study treatment due to disease progression or relapse, and the overall global composite score was determined at that time, even if not all compartments had been assessed. If clinical progression was noted at any time before the scheduled assessments for efficacy, the assessments (CT, mSWAT, skin photographs, and flow cytometry) were done at that time to fully document disease progression.²⁷

As per the study’s published global composite response criteria, in cases where the definition of PD or relapse was met but the clinical impression was questionable, the patient could remain on study for at least 4 weeks to avoid them being removed prematurely from the study. Clinical progression was considered for patients with disease progression that did not meet the criteria for PD based on CTCL response criteria. If PD was confirmed at the subsequent evaluation, the patient was to be discontinued from treatment. If PD was not confirmed at the subsequent evaluation, the patient could remain on study. If patients experienced an overall CR, treatment could continue for up to 12 months or until PD, whichever came first.²⁷

Additionally, a blinded independent review of data for each patient was performed by an independent review committee to determine progression and date of progression. The independent reviewer was supplied with the following: results of the independent radiology review (assessment of lymph nodes and viscera), mSWAT score as assessed by the investigator, results of central flow cytometry, results of biopsies if appropriate for assessment of disease, and date of death for patients who died before progression. The independent reviewer also had access to CT scans and skin photographs. The reviewer also reviewed the date of any objective responses (CR or PR).²⁷

Health-Related Quality of Life

The change from baseline through 6 months for the Skindex-29 score, FACT-G total score, and EQ-5D index score was also a key secondary end point of the MAVORIC trial.²⁷ The Skindex-29, FACT-G, and EQ-5D-3L were evaluated at baseline (cycle 1, day 1) and at the end of cycle 1, then every 8 weeks through cycle 25 (i.e., end of cycle 3, 5, and up to cycle 25) and at the end-of-treatment visit.³⁰

A detailed discussion and critical appraisal of the HRQoL outcome measures is provided in Appendix 4:

• The Skindex-29 is a validated instrument to measure the effect of skin disease on HRQoL composed of 29 items assessing 3 domains: emotions, symptoms, and functioning on a 5-point Likert scale. Higher scores indicate a higher impact of skin disease. Responses to each item are transformed to a linear scale of 100 (never = 0, rarely = 25, sometimes = 50, often = 75, all the time = 100) for the purpose of calculating a scale score. A scale score is the mean of a patient’s responses to the items in each scale, and the composite Skindex-29 score is calculated as the average of the 3 scale scores to measure the overall impact on QoL. Higher scores indicate a higher impact of skin disease. Missing data for Skindex-29 were handled according to the questionnaire’s scoring guidelines. If responses to more than 25% of items were missing overall, the Skindex-29 score was treated as missing. If any scale had more than 25% of the responses missing, the scale score was missing. The scale scores were the average of the non-missing items in a given scale.²⁷ The recall period of the Skindex-29 was 1 week.³⁰

• The FACT-G is a validated instrument for assessing HRQoL in patients with cancer consisting of 27 items in 4 domains: physical well-being, social/family well-being, emotional well-being, and functional well-being. The total FACT-G score is obtained by summing individual subscale scores. Response scores on negatively phrased questions are reversed before summing. Higher scores for the scales and subscales indicate better QoL. Missing data for FACT-G were handled according to the questionnaire’s scoring guidelines. If more than 50% of the items for a subscale had responses, the subscale scores were prorated by multiplying the sum of the subscale by the number of items in the subscale and then dividing by the number of items actually responded to. The total FACT-G score was scored only if the overall item response rate was greater than 80%.²⁷ The recall period of the FACT-G was 7 days.³⁰

• The EQ-5D is a standardized, reliable, and validated instrument to measure HRQoL. The EQ-5D self-reported questionnaire includes the EQ-5D descriptive system and VAS. The 3-level version (EQ-5D-3L) descriptive system comprises the following dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. The EQ-5D index score is calculated based on the descriptive system using a set of item weights (value sets) to derive a single score ranging from −0.109 to 1, with 1 representing full health. The VAS records the patient’s self-rated health status on a scale of 0 to 100, with higher scores indicating better health. For EQ-5D-3L, the EQ-5D index score was treated as missing if not all 5 descriptors were responded.²⁷ The recall period of the EQ-5D-3L was the current period.³⁰

Other Secondary Efficacy Outcomes

Other secondary efficacy outcomes included PFS and ORR by independent review, BOR, DOR, changes in HRQoL measures at other time points, and change in pruritus evaluation. Secondary outcomes were assessed by both the investigator and independent review. Secondary outcomes in the MAVORIC trial were defined as follows:²⁷

• BOR: Defined as the best response recorded across all time points from the start of treatment until disease progression/recurrence or end of treatment.

• DOR: Defined as the time from the date that criteria for CR or PR (whichever was recorded first) were met until the first date that PD or death was objectively documented.

• Change in pruritus evaluation: Change in pruritus was evaluated by the Pruritus Likert Scale and the ItchyQoL questionnaire. The Pruritus Likert Scale measures the level of itching for pruritus on a scale of 0 to 10, with higher scores indicating greater itch. The ItchyQoL is a validated pruritus-specific QoL instrument consisting of 22 pruritus-specific questions in 3 major domains: symptoms, functioning, and emotions. Higher ItchyQoL scores indicate worse QoL. The ItchyQoL and the Pruritus Likert Scale were evaluated every 4 weeks at each cycle.³⁰ The recall period for the ItchyQoL was not specified.

Exploratory Efficacy Outcomes

Exploratory efficacy end points included OS, defined as the time from the date of randomization until the date of death due to any cause, and TTF, defined as the time from the day of randomization until discontinuation of randomized treatment due to any reason, except for those who discontinued randomized treatment due to 1 year on treatment with a CR. Patients who were still alive at the end of the survival follow-up period or were lost to follow-up were censored on the last date the patient was known to be alive.²⁷

An additional exploratory analysis was performed for comparison of PFS between patients with any exposure to mogamulizumab versus vorinostat. For this analysis, the PFS for patients who cross over from the vorinostat arm to the mogamulizumab arm was calculated from the first dose of mogamulizumab.³⁵

Safety end points included AEs defined as any untoward medical occurrence in a patient in the clinical trial who had been administered a study drug (i.e., investigational product) or a comparator product. The intensity of AEs was graded using the National Cancer Institute’s Common Terminology Criteria for Adverse Events v.4.0. In the randomized treatment period, TEAEs were defined as AEs that occurred from the first dose of the randomized study drug through 90 days after the last dose of randomized study drug or the start of alternative CTCL therapy, whichever occurred first. The study protocol specified that disease progression was not to be considered an AE. Lymphopenia, the pharmacologic effect of the drug, was also not to be considered an AE. However, in cases where the investigators did record disease progression or lymphopenia as AEs, these events were included in the reporting and analysis of AEs.²⁷

For patients who crossed over from vorinostat to mogamulizumab, TEAEs during the randomized treatment period were defined as AEs that occurred after the first dose of vorinostat and before the first dose of mogamulizumab. AEs that occurred after the start of mogamulizumab but were considered related to vorinostat were also counted as TEAEs for the randomized treatment period. The incidence of specific TEAEs in patients who crossed over was not reported.²⁷

Statistical Analysis

Sample Size and Power Calculation

The sample size calculation for the MAVORIC study was based on the primary end point of PFS in the ITT population. Assuming a median PFS for vorinostat of 169 days for reference, a 50% improvement in the median PFS for mogamulizumab was predicted, resulting in a predicted median PFS for mogamulizumab of 254 days. Based on a 24-month accrual and 12-month follow-up, and a 10% rate for dropouts and sample size inflation, the estimated sample size was 317 patients. The sample size of 317 patients provided the 90% power necessary to observe 255 PFS events at a 1-sided 0.025 significance level.²⁷

The final primary analysis comparing PFS between treatment groups was to be conducted when a total of 255 PFS events had occurred or a maximum of 24 months after the last randomized patient’s first dose, whichever came first. In the event the study was stopped before 255 PFS events being observed, the primary test was to be performed at less than 90% power under the current assumptions, according to the protocol.²⁷

The planned enrolment and 255-event threshold for the primary analysis assumed that approximately 20% of the 317 randomized patients would be censored for the primary end point at the time of the primary analysis. Additionally, the 24-month time threshold for the primary analysis assumed that meaningful numbers of PFS events would accrue during the 24-month period after the enrolment of the last patient.²⁷

Interim and Final Analyses

No formal interim analyses were planned or conducted for the MAVORIC study. The final primary analysis was conducted after 255 PFS events or a maximum of 24 months after the last randomized patient’s first dose.²⁷

Analysis Populations

The following analysis populations were defined in the MAVORIC study:²⁷

• ITT set: Includes all patients randomized to a therapy (mogamulizumab or vorinostat) and assigned a study number.

• Safety analysis set: Includes all patients who received at least 1 dose (even a partial dose) of the assigned study drug (mogamulizumab or vorinostat; does not include patients who crossed over from vorinostat to mogamulizumab).

• Efficacy evaluable set: Includes all patients who received the first cycle of treatment (at least 1 dose) and who had a baseline tumour assessment and at least 1 post-baseline assessment for response. If a patient had a baseline tumour assessment but progressed (either died before progression or had documented progression in the follow-up period) during the study without any post-baseline tumour assessment, that patient was still considered as part of the efficacy evaluable set.

Efficacy Analyses

Summary statistics were presented by treatment arm. For continuous variables, the number of available observations (n), mean, standard deviation (SD), median, minimum, and maximum were provided. For categorical variables, the frequency and percentage in each category were displayed. Additionally, the point estimates were accompanied by the corresponding 2-sided 95% CIs.²⁷

Primary Efficacy End Point

The primary efficacy end point of the MAVORIC trial was PFS per investigator assessment, performed on the ITT population using a stratified log-rank test at the 1-sided 2.5% significance level. Stratification was by disease type (MF or SS), disease stage (IB or II versus III or IV), and region (US, Japan, and rest of world). A Cox proportional hazard model with treatment, disease type, disease stage, and region (US, Japan, and rest of world) as covariates was used to assess the magnitude of the treatment difference in PFS. The hazard ratio (HR) along with the 95% CI obtained from the Cox proportional hazard model were presented. The median PFS and the 2-sided 95% CI for each treatment were estimated using the Kaplan-Meier method. The Kaplan-Meier estimate of PFS rates and the corresponding 95% CIs were also provided for each treatment arm by 6-month interval. Plots of the Kaplan-Meier estimate of the survival distribution function for PFS over time were presented by treatment arm.²⁷

An additional analysis of the treatment effect on PFS when adjusted for potential prognostic factors was carried out using a multivariate Cox proportional hazard model. Analyses were performed based on PFS as assessed by the investigator for the first assigned therapy and the ITT set. Potential prognostic factors including disease type (MF or SS), disease stage (IB or II versus III or IV), compartment involvement (blood involvement or no blood involvement), region (US, Japan, rest of world), age group (< 65 or ≥ 65 years), gender (males or females), and race category (Black or African American, White, Other) were included as covariates in the multivariate Cox proportional hazard model for PFS. Backward selection was used to identify the final set of prognostic factors (exit P value was set to 0.1). The final model was then used to assess treatment effect when adjusted for these important prognostic factors.²⁷

Sensitivity Analyses

A total of 4 sensitivity analyses using different definitions of PFS were conducted for the primary end point in the ITT population based on investigator assessment:²⁷

• This analysis defined PFS as the time from the day of randomization until documented progression in any compartment, based on the investigator’s assessment per CTCL response criteria or death due to any cause, provided that death was not more than 8 weeks after the last post-baseline tumour assessment or the last dose of the study drug (if the patient did not have any post-baseline tumour assessments). Using this definition, the progression reported during the follow-up period would not be considered an event in the sensitivity analysis but would be a censored value on the day of the last tumour assessment.

• This analysis defined PFS as the time from the day of randomization until the earliest date of documented progression in any compartment based on the investigator’s assessment per CTCL response criteria, or clinical progression at the end of the randomized treatment period, or death due to any cause. Using this definition, clinical progression noted at the end of the randomized treatment would be considered an event in the sensitivity analyses.

• This analysis defined PFS as the time from the day of randomization until documented progression or death due to any cause. The date of documented disease progression was the earliest date of disease progression in any compartment based on the investigator’s assessment per CTCL response criteria or disease progression reported during the follow-up period. For assessments where progression was reported by the investigator but was not confirmed by the independent review, the progression date was set to the last tumour assessment plus 1 day in the mogamulizumab treatment arm and was censored in the vorinostat treatment arm.

• This analysis defined PFS as the time from the day of randomization until documented progression in any compartment per CTCL response criteria or death due to any cause, provided that death occurred:

  • within 90 days after the last dose of the study drug for patients who had no post-baseline tumour assessments

  • within 56 days of the last post-baseline tumour assessment if that date was more than 90 days after the last dose of study drug

  • within 90 days of the last dose of the study drug if the date corresponding to 56 days after the last post-baseline tumour assessment was 90 days or more after the last dose of the study drug.

Subgroup Analyses

Pre-specified subgroup analyses for the primary end point of PFS in the MAVORIC trial were conducted on the ITT population and efficacy evaluable set and included disease type (MF versus SS), disease stage (IB or II versus III or IV), blood involvement (yes versus no), region (US versus Japan versus rest of world), age (< 65 years versus ≥ 65 years), gender (male versus female), race (Black or African American versus White versus other), and lactate dehydrogenase (normal versus elevated). Results for subgroup analyses were assessed by the investigator in the ITT population and presented as forest plots of HRs.²⁷

Key Secondary Efficacy End Points

Objective Response Rate

ORR was a key secondary end point of the MAVORIC study. A confirmed CR or PR was defined as a documented CR or PR based on the investigator’s assessment of overall response per a global composite response score that was subsequently confirmed by 2 or more consecutive observations for a minimum of 4 weeks. Mogamulizumab and vorinostat were compared using a Cochran-Mantel-Haenszel test adjusted for disease type, disease stage, and region). The exact 95% CIs for ORR were calculated for each treatment arm along with the difference in response rates between the 2 treatment arms. The number and percentage of patients achieving objective response (CR or PR) were presented for each treatment arm.²⁷

Additional analysis of the treatment effect on ORR was conducted as previously described for PFS; however, a multivariate logistic regression model was used for ORR. Potential prognostic factors were the same for the PFS and ORR end points.²⁷

Crossover from the vorinostat arm to the mogamulizumab arm was permitted in patients who had received 2 full treatment cycles and demonstrated progression of disease at the 8-week (cycle 2, day 26 to 28) assessment, or anytime thereafter. For patients who crossed over from the vorinostat arm to the mogamulizumab arm, ORR was estimated in a similar fashion as described previously. All summaries for the crossover portion were purely descriptive and no statistical comparisons were made.²⁷

Health-Related Quality of Life

HRQoL, as measured by the Skindex-29, FACT-G, and EQ-5D, were key secondary end points of the MAVORIC study and were evaluated in the ITT and efficacy evaluable populations.²⁷ An additional analysis of the HRQoL data was submitted that evaluated HRQoL as measured by the Skindex-29, FACT-G, and EQ-5D (and ItchyQoL and Pruritis Likert scale) in the cross-sectional analysis population that was derived from the MAVORIC ITT, which was defined at each QoL collection time point as any patient who had survived and maintained in the randomized period of the study into day 1 of each cycle and had available QoL data for that cycle. The longitudinal-period population was defined as all patients with an outcome collection period reported from baseline to each cycle for analysis. Any patient who had survived and maintained in the study from baseline to day 1 of cycle 1 to day 1 of a specified cumulative cycle (e.g., cycle 2), or who had discontinued treatment and had available QoL data for that time point, were included.³⁰

For all HRQoL measures, the mean change from baseline in the Skindex-29 score, FACT-G total score, and EQ-5D-3L index score was evaluated. Additionally, for the Skindex-29, the mean between-group difference was assessed. Assessments of HRQoL were conducted at the end of cycle 1, cycle 3, and cycle 5, up to 6 months, or cycle 11. (Data beyond cycle 11 were summarized descriptively). A restricted maximum likelihood method was used as well as a repeated measures mixed-effects model with time, region, disease type, disease stage, treatment, time-by-treatment interaction as factors, and baseline score was a covariate. An unstructured covariance matrix was employed to model the correlation among repeated measurements. Appropriate contrasts were used to determine the difference between treatment groups across the end of cycles 1, 3, and 5. In the event of convergence problems, another variance-covariance structure was to be considered. The selection of any of these structures was determined after exploration of the observed correlation structure. Least squares mean, corresponding standard errors, and 2-sided 95% CIs were presented for the within-group change. For the between–treatment group comparison, the difference in least squares mean, corresponding standard errors, 2-sided 95% CIs, and 2-sided P value were also derived from this analysis of covariance (ANCOVA) model. This mixed-effects model was also used to compare the treatment differences in the mean changes in Skindex-29 score from baseline to each scheduled time point at the end of cycle 1, cycle 3, and cycle 5. If residuals of the ANCOVA model were not normally distributed, then the analysis was carried out using a Wilcoxon rank sum test instead. In addition to the analysis described previously, summary statistics, including n, mean, SD, median, minimum, and maximum, were provided for the Skindex-29, FACT-G, and EQ-5D scores at baseline and each scheduled time point, and the change in the total score at each of these scheduled assessments for each treatment arm.²⁷

Multiplicity Adjustment

For the 4 key secondary end points (i.e., ORR, and change in the Skindex-29 score, FACT-G total score, and EQ-5D-3L index score from baseline through the 6-month assessment), comparisons between the 2 treatment arms were conducted using P values that were adjusted to control the overall studywise type I error rate to be less than 0.05. Since these tests were not independent, the 4 key secondary end point analyses were conducted using the Šídák adjusted P value method:²⁷

Adjusted P value = 1 − (1 − P),⁴ where P is the original P value of the individual test.

These adjusted P values were then compared to 0.05 for each test.²⁷

Other Secondary Efficacy End Points

Independent Review Assessment of PFS and ORR

Progression-free survival and ORR were also assessed by independent review using the same methods described previously for PFS and ORR.²⁷

Best Overall Response

BOR was a secondary outcome of the MAVORIC trial. BOR was assessed by the investigator and by the independent reviewer during the randomized treatment period for all patients by treatment arm. BOR was summarized using frequency tables, and concordance between the investigator’s assessment and the independent review assessment of BOR was summarized.²⁷

Duration of Response

DOR was a secondary end point of the MAVORIC trial. No comparisons between mogamulizumab and vorinostat were made, as DOR is conditional on patients achieving a confirmed response. The DOR was analyzed using survival analysis methods. The Kaplan-Meier estimate of the median DOR and the associated 95% CIs were estimated and presented separately for each treatment arm.²⁷

Time to Response

Time to response was a secondary outcome of the MAVORIC trial. For patients who achieved a BOR of CR or PR during the randomized treatment period, the time to response was summarized descriptively. Time to response (TTR) was defined as the time from the date of randomization to the date that criteria for CR/PR (whichever was first recorded) were first met. Subjects who did not respond over the course of the study had a missing value for TTR.²⁷

Changes in Pruritus Evaluation

The Likert scale for pruritus evaluation uses a numbered scale from 0 to 10 to measure the level of itching for pruritus with 10 indicating the worst itch imaginable and 0 indicating no itch. For the Pruritus Likert Scale data, the changes in the Likert scale score from baseline at different time points up to cycle 11 assessments during the randomized treatment period were analyzed using the Wilcoxon rank sum test. The treatment difference in the median change of the Likert scale score from baseline was estimated using the Hodges-Lehmann estimate, and a 2-sided 95% CI for the treatment difference was obtained using the Moses method.²⁷

The ItchyQoL includes 22 pruritus-specific questions covering 3 major domains: symptoms, functioning, and emotions. The subscale scores consist of the average of the responses to the items in a given subscale. The overall score is the average of the responses to all items. Higher ItchyQoL scores indicate worse QoL. Missing data in the ItchyQoL questionnaire were handled according to the following rule: if responses to more than 25% of items were missing overall, the overall score was treated as missing; if any scale had more than 25% of the responses missing, the scale score was missing.²⁷

Exploratory Efficacy End Points

Overall Survival

The OS rates for the mogamulizumab and vorinostat arms were compared as described for PFS based on the ITT population; however, this analysis was considered exploratory. Patients who were still alive at the end of the survival follow-up period or were lost to follow-up at the time of the analysis were censored on the last date the patient was known to be alive.²⁷

Additional exploratory analyses of OS adjusting for the 1-way crossover were performed, which included censoring patients at the time of crossover, as well as the rank-preserving structural failure time (RPSFT) model, and the inverse probability censoring weighting (IPCW) method. The RPSFT model assumes that the experimental treatment is acting by multiplying survival time by a given factor once the patient starts receiving the experimental treatment. The RPSFT model reconstructs the survival time for the control arm as if no crossover were allowed and analyzes data based on reconstructed data. The IPCW method censors the patients who crossed over at the time of crossover, but patients were weighted according to their probability of crossover. The IPCW method artificially increases weights for patients with a low probability of crossover and decreases weights for patients with a high probability of crossover.³⁵

Time-to-Treatment Failure

TTF for the mogamulizumab and vorinostat arms was compared as described for PFS based on the ITT population; however, it was considered exploratory. Patients who experienced a CR and discontinued randomized treatment after 1 year of treatment were censored at the last dose date of the randomized treatment. Patients who were randomized but did not take any study drug were censored at the last documented visit date.²⁷

Safety Analyses

All safety summaries were based on the safety analysis set. TEAEs during the randomized treatment period were summarized by the number and percent of patients in each treatment arm and in total by SOC and preferred term for the safety population, and subgroups of gender (male versus female) and age (< 65 years versus ≥ 65 years).²⁷

If a patient who had received mogamulizumab underwent a stem-cell transplant at any time during or after the 90-day follow-up period, AEs and SAEs were to be documented from the transplant procedure through 180 days post transplant.²⁷

The safety data for patients in the vorinostat arm who crossed over to mogamulizumab were summarized separately in the same manner as those in the randomized treatment groups and presented separately. No safety results for patients who crossed over were provided.²⁷

Protocol Amendments and Deviations

The sponsor-submitted Clinical Study Report included a total of 8 amendments that were made to the original protocol (issued June 19, 2012) of the MAVORIC study, which are summarized in Table 13. The submitted protocol corresponded to amendment 10, occurring on May 31, 2018. The first protocol amendment was conducted in July 2012, following requests from the FDA. The order of secondary objectives was modified at amendment 2 to reflect the importance of QoL data and to specify the exploratory objectives of OS and mogamulizumab exposure–response relationships.²⁷ Most of the major protocol changes affected the inclusion and exclusion criteria of the patients eligible for the study. Amendment 5, which occurred 2 years after the original protocol, clarified that patients with either relapsed or refractory disease were eligible to participate, potentially impacting the type of patients included.²⁷ Other protocol amendments were generally related to changes in or clarification of study procedures.

The last patient was randomized on January 29, 2016; however, it is unclear how many patients were enrolled before and after each amendment; therefore, it is difficult to determine any potential impacts the protocol changes may have had on the overall population. However, the changes to the protocol were not believed to impact the conduct or integrity of the study, nor did they affect the patients’ safety.

There was a total of 99 patients (53.2%) in the mogamulizumab group and 111 patients (59.7%) in the vorinostat group with major protocol deviations in the MAVORIC trial (Table 14).²⁷

Changes to Planned Analysis

The original statistical analysis plan was signed on February 24, 2014, and revised on November 22, 2016. Key changes to analyses that were not specified in the statistical analysis plan before database lock include the following:²⁷

• clarified that for PFS, documented progression includes disease progression in any compartment based on investigator’s assessment per CTCL response criteria, or documented disease progression reported during the follow-up period

• clarified that confirmation of a CR and PR was required for the ORR and that confirmed and unconfirmed ORRs would be summarized; added summaries by disease type

• added TTF as an exploratory efficacy variable

• removed statement that each patient would be followed in the study for a minimum of 2 years to generate information on OS

• added a sensitivity analysis for PFS that included clinical progression as a PFS event

• added exploratory analysis of PFS for patients with any exposure to mogamulizumab versus patients who received vorinostat only

• removed the CCR4-related analysis (a separate analysis plan was prepared to detail the CCR4 analysis)

• added testing for QoL assessments beyond the 6-month assessment period

• added exploratory analyses of OS adjusting for crossover; these analyses included the RPSFT model and the IPCW model.

Minor changes to the planned analyses defined in the final revised statistical analysis plan were provided in a supplement dated March 21, 2017 (before database lock):²⁷

• added a fourth sensitivity analysis of PFS to consider death during safety follow-up

• clarified that exploratory analyses of OS accounting for crossover effect (RPSFT and IPCW analysis models) would not be presented in the Clinical Study Report but would be included elsewhere in the Common Technical Document (module 2.7.3)

• specified that AEs that occurred more than 90 days after the last dose of the study drug but were related to the study drug would be considered as TEAEs.

Subsequent to database lock and availability of topline results, the following additional analyses were performed:²⁷

• summary of predefined protocol deviations

• analysis and summary of dose intensity for vorinostat

• analyses of response by compartment (blood, lymph nodes, skin, viscera):

  • confirmed response rate based on investigator’s assessment and independent review

  • confirmed plus unconfirmed response rate based on investigator’s assessment and independent review

  • DOR based on investigator’s assessment and independent review

  • BOR based on investigator’s assessment and independent review

  • TTR based on investigator’s assessment and independent review

• confirmed ORR by number of prior systemic therapies received based on investigator’s assessment and independent review

• summary of PFS based on investigator’s assessment, excluding patients who had specific protocol deviations.

Results

Patient Disposition

Table 15 summarizes the disposition of patients enrolled in the MAVORIC trial. The MAVORIC trial was a randomized, open-label, phase III clinical trial. A total of 464 patients were screened for eligibility, and 372 were randomized to receive mogamulizumab (n = 186), or vorinostat (n = 186), making up the ITT set, for which analyses did not include patients who crossed over from vorinostat to mogamulizumab. Two patients randomized to the mogamulizumab arm did not receive the study treatment (n = 184), comprising the safety analysis set, which also did not include patients who had crossed over. As of the data cut-off date (December 31, 2016), 27 (14.5%) and 10 (5.4%) randomized patients remained on mogamulizumab and vorinostat, respectively. The most common reason for discontinuation of randomized treatment in the mogamulizumab group was PD per CTCL response criteria in any category of skin, nodes, blood, and viscera (n = 76 out of 157; 48.4%).²⁷

At the time of data cut-off, 109 patients (58.6%) randomized to mogamulizumab were continuing the study (treatment or follow-up), and 77 (41.4%) had discontinued the study, primarily due to death (n = 40; 21.5%) or withdrawal of consent (n = 30; 16.1%). Of the 186 patients randomized to vorinostat, 70 (37.6%) had discontinued from the study at the time of data cut-off, including 41 who had crossed over to mogamulizumab treatment, primarily due to death (n = 47; 25.3%) or withdrawal of consent (n = 19; 10.2%).²⁷

A total of 136 patients (73.1%) randomized to the vorinostat group crossed over to receive treatment with mogamulizumab, while 40 patients randomized to vorinostat did not cross over but discontinued vorinostat, and 10 patients remained on randomized treatment. The most common reason for crossover was due to disease progression (n = 109; 80.1%), followed by drug intolerance (n = 27; 19.9%). At the time of data cut-off, 105 patients (77.2%) who crossed over from vorinostat had discontinued study treatment. The primary reasons for discontinuation following crossover were disease progression per CTCL response criteria or clinical disease progression (n = 61; 44.9%) and AEs (n = 22; 16.2%).²⁷ Of the 61 patients who crossed over to mogamulizumab and discontinued due to progression, 46 experienced at least stable disease before progressing on mogamulizumab, 13 continued to progress, and 2 patients approved for crossover did not receive mogamulizumab treatment before progression.⁴⁴ Thirty-one patients (22.7%) who crossed over from vorinostat were continuing to receive mogamulizumab at data cut-off.²⁷

The last patient exited the MAVORIC trial on February 17, 2021.⁴⁴

Exposure to Study Treatments

Exposure to study treatments is summarized in Table 16. The median follow-up of the randomized phase of the MAVORIC trial was 17.0 months for efficacy and 34.5 months for safety.⁴³ The mean and median duration of exposure to study treatments was higher in the randomized mogamulizumab arm and in the patients who crossed over to mogamulizumab compared with vorinostat. The median duration of treatment with mogamulizumab was 170 days compared with 84 days with vorinostat in the randomized treatment period. The median dose intensity was 97.49% and 95.12% for the mogamulizumab and vorinostat groups, respectively.

Results were similar in patients who crossed over to mogamulizumab from vorinostat with a similar mean number of cycles of mogamulizumab administered between randomized and crossover patients; the median duration of treatment was 169 days and the median dose intensity was 94.92%. Additionally, the proportion of patients initiating each cycle was similar between the patients randomized to mogamulizumab and those who crossed over to mogamulizumab.²⁷

Dose Interruptions and Modifications

During the randomized treatment period, 65 patients (35.3%) in the mogamulizumab group had at least 1 dose withheld, while 70 (38.0%) did not receive the total planned dose of mogamulizumab. Mogamulizumab administration was temporarily interrupted in 17 patients (9.2%), primarily due to infusion reaction. Conversely, 101 patients (54.3%) in the vorinostat group had at least 1 dose modification during the randomized treatment period, and 34 patients (18.3%) missed at least 1 dose.²⁷

In the crossover portion of the trial, 38 patients (27.9%) who crossed over to receive mogamulizumab had a dose withheld, 43 (31.6%) did not have their total planned dose administered, and 17 (12.5%) had their infusion temporarily interrupted.²⁷

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported subsequently. See Appendix 3 for detailed efficacy data.

Overall Response Rate

The ORR in the ITT population as assessed by the investigator was a key secondary efficacy end point of the MAVORIC trial. Results for ORR are summarized in Table 17. A total of 28% (95% CI, 21.6 to 35.0) of patients in the mogamulizumab group and 4.8% (95% CI, 2.2 to 9.0) of patients in the vorinostat group achieved a confirmed ORR. The risk difference between groups was 23.1%; P < 0.0001). Including both confirmed and unconfirmed responses, ORR based on investigator assessment was 34.9% for mogamulizumab compared with 6.5% for vorinostat (P < 0.0001).²⁷

ORR by Independent Review

The ORR assessed by independent review was a secondary end point of the MAVORIC study. Results for confirmed ORR based on independent review were consistent with ORR based on investigator assessment. The confirmed ORR by independent review was 23.1% for mogamulizumab compared with 3.8% for vorinostat (P < 0.0001). The risk difference between mogamulizumab and vorinostat as assessed by independent review was 19.4% (95% CI, 9.0 to 29.4).²⁷

Additional (Subgroup) Analysis of ORR

Multiple unplanned subgroup and post hoc analyses of ORR were identified for the MAVORIC trial, including analyses by disease type (MF versus SS),²⁷ disease stage,^28,45 disease compartment,^27,45 blood involvement classification at baseline,²⁹ number of prior systemic therapies,²⁷ and type of prior systemic therapy.⁴⁶

Though not an official pre-specified subgroup analysis in MAVORIC, ORR was estimated by disease type for patients with MF or SS at baseline. The results by disease type were consistent with the primary ORR analysis, where the proportion of MF patients with ORR was 21.0% in the mogamulizumab group compared with 7.1% in the vorinostat group, and 37.0% versus 2.3% in the mogamulizumab and vorinostat groups for patients with SS.²⁷

A post hoc analysis conducted on patients from the MAVORIC trial with less advanced (stage IB and IIA) MF (n = 36 for mogamulizumab; n = 49 for vorinostat) demonstrated consistent ORR results with the primary analysis, with an ORR of 19.5% for mogamulizumab (20% in patients with stage IB and 19% in patients with IIA), compared with 10.2% for vorinostat (18.5% in IB and 0% in IIA). For patients with stage IB and IIA disease, compartment response rates for blood, skin, and lymph node compartments were 75%, 38.1%, and 15.0% for mogamulizumab, and 12.5%, 22.7%, and 0% for vorinostat, respectively.⁴⁵ Conversely, in another post hoc analysis of patients with advanced (stage IIB to IVB) MF or SS (mogamulizumab, n = 150; vorinostat, n = 137), the investigator-assessed ORR was 30.0% for mogamulizumab and 2.9% for vorinostat.²⁸

A post hoc analysis assessing response rate by disease compartment was performed to assist in the interpretation of the key secondary end point of ORR (Table 18). Patients were included in the analysis for a specific compartment if disease was present in the compartment at baseline or the patient had a post-baseline assessment indicating the presence or progression of disease in a compartment not involved at baseline. The investigator-assessed ORR in skin was 41.9% for mogamulizumab and 15.6% for vorinostat. The response rate in blood was also higher for mogamulizumab, with an investigator-assessed ORR of 66.9% for mogamulizumab and 18.4% for vorinostat. The ORR in lymph nodes was 15.4% for mogamulizumab compared with 3.8% for vorinostat. There were no confirmed responses achieved in visceral involvement in either arm as assessed by the investigator. Similar results were observed for ORR by compartment based on independent review (Table 40), though the proportion of patients with ORR in blood and lymph nodes was more than 5% higher, as assessed by the investigator, compared with independent review.²⁷

The results for ORR by blood classification in a post hoc analysis are summarized in Table 45 of Appendix 3. When stratified by baseline blood classification, the ORR appeared higher with mogamulizumab than with vorinostat in all B₀ (15.6% versus 6.5%), B₁ (25.8% versus 6.5%), and B₂ (37.4% versus 3.2%) assessments.²⁹

An additional analysis subsequent to the main ORR results was the assessment of ORR by number of prior systemic therapies. The response rate for mogamulizumab was greater than that for vorinostat, regardless of the number of prior systemic regimens received, as assessed by the investigator (Table 19) and independent review (Table 41).²⁷

ORR by prior type of therapy received in patients treated with mogamulizumab is summarized in Table 47 and Figure 11 of Appendix 3. The ORR was relatively consistent, with the primary analysis ranging from 20.4% for immunoinhibitory treatments to 38.5% for HDAC inhibitors, and it was also consistent by specific systemic therapy, ranging from 19.6% for oral bexarotene to 37.5% for romidepsin.⁴⁶

Best Overall Response

BOR was a secondary efficacy end point of the MAVORIC study. It was defined as the best response recorded across all time points from the start of treatment until disease progression or recurrence or end of treatment. Results for BOR as assessed by the investigator and independent review were consistent and are summarized in Table 20.²⁷

Overall, there were more responders in the mogamulizumab group compared with the vorinostat group, as assessed by both the investigator (34.9% versus 6.5%) and independent review (29.0% versus 7.0%). As reported by the investigator, there were 4 confirmed CRs in the mogamulizumab group and none in the vorinostat group. The rate of confirmed PR as assessed by the investigator was higher in the mogamulizumab group compared with the vorinostat group (25.3% versus 4.8%); however, the proportion of patients with stable disease was higher in the vorinostat group compared with mogamulizumab (61.8% versus 43.0%). Results by independent review were consistent with the investigator’s assessment.²⁷

Of the 136 patients who were randomized to vorinostat and crossed over to mogamulizumab, responses to mogamulizumab were observed in 41 patients (30.1%) during the crossover portion of the study. There were 7 (5.1%) confirmed CRs and 34 (25.0%) confirmed PRs that occurred during crossover.²⁷

Duration of Response

DOR for patients with a confirmed CR or PR is summarized in Table 21. The median DOR as assessed by the investigator for confirmed responders was 14.07 months (95% CI, 9.43 to 19.17) for mogamulizumab and 9.13 months (95% CI, 4.67 to NE) for vorinostat. Results for DOR were consistent when assessed by independent review (Table 43); however, the median DOR could not be assessed for vorinostat, as 6 of the 7 observations were censored.²⁷

DOR by compartment was assessed as a post hoc analysis. The median duration of confirmed responses by investigator assessment in skin was 20.57 months (95% CI, 17.77 to NE) for mogamulizumab and 10.67 months (95% CI, 4.77 to NE) for vorinostat. The median duration of confirmed responses in blood and nodal responses for mogamulizumab was 25.47 months (95% CI, 16.10 to NE), and 15.47 months (95% CI, NE to NE). The median duration of confirmed responses for vorinostat in these compartments was not evaluable. The duration of confirmed responses was not evaluated in the visceral compartment, as no responses were observed.²⁷

Additional (Subgroup) Analysis of DOR

In a post hoc analysis, the DOR was measured by number and type of prior treatment for patients in the MAVORIC trial. The median DOR by therapeutic class is summarized in Table 47, while the median DOR by specific prior therapy is summarized Figure 11 of Appendix 3. Median DOR in patients treated with mogamulizumab in the MAVORIC trial was consistent with the primary analysis, regardless of the number of prior therapies received, ranging from 9.4 months with 6 or more prior therapies to 19.9 months with 4 prior therapies (Table 46). Additionally, the DOR was consistent regardless of therapeutic class (Table 47), or specific therapy received (Figure 11).⁴⁶

Progression-Free Survival

PFS per investigator assessment was the primary end point of the MAVORIC trial. Results for PFS are summarized in Table 22 and by Kaplan-Meier plot in Figure 2. At the time of the data cut-off (December 31, 2016), a total of 241 PFS events had occurred; 110 (59.1%) in the mogamulizumab group and 131 (70.4%) in the vorinostat group.²⁷ The HR for mogamulizumab versus vorinostat was 0.53 (95% CI, 0.41 to 0.69), in favour of mogamulizumab. The median PFS was 7.70 months (95% CI, 5.67 to 10.33) in the mogamulizumab group, and 3.10 months (95% CI, 2.87 to 4.07) in the vorinostat group (P < 0.0001). Patients without progression who were continuing to receive randomized treatment were censored as of the cut-off date. The PFS rate at 6, 12, 18, and 24 months was 55.3% (95% CI, 47.1 to 62.6), 38.3% (95% CI, 30.2 to 46.4), 28.0% (95% CI, 19.8 to 36.8), and 14.1% (95% CI, 6.4 to 24.8) in the mogamulizumab group, and 28.8% (95% CI, 21.6 to 36.3), 15.3% (95% CI, 9.5 to 22.3), 7.2% (95% CI, 2.7 to 14.5), and 7.2% (95% CI, 2.7 to 14.5) for the vorinostat groups, respectively.²⁷

A total of 133 patients crossed over from vorinostat to receive treatment with mogamulizumab due to either disease progression or intolerance. In the pre-specified additional exploratory analysis of PFS for these patients (i.e., PFS2), the median PFS2 calculated from the first dose of mogamulizumab was 8.87 months (95% CI, 5.37 to 14.77). Across all patients who were either randomized to mogamulizumab or received mogamulizumab during crossover (n = 319), the median PFS was 8.43 months (95% CI, 6.10 to 10.30).²⁷

Figure 2: Kaplan-Meier Curve of PFS by Investigator’s Assessment (ITT)

ITT = intention-to-treat population; KW = KW-0761 (mogamulizumab); PFS = progression-free survival; VOR = vorinostat.

Source: MAVORIC (0761 to 010) Clinical Study Report.²⁷

Sensitivity Analysis of PFS

Results of sensitivity analyses for PFS were consistent with the primary analysis and are summarized in Table 44 of Appendix 3. In all cases, the PFS was favoured for the mogamulizumab group, with HRs of 0.52 (95% CI, 0.40 to 0.68), 0.61 (95% CI, 0.47 to 0.78), 0.72 (95% CI, 0.54 to 0.96), and 0.52 (95% CI, 0.40 to 0.68) for sensitivity analyses 1, 2, 3, and 4, respectively.²⁷

In addition to the pre-specified sensitivity analyses, an ad hoc analysis of PFS was conducted that excluded patients with protocol deviations identified by the sponsor as potentially affecting the efficacy results. The results of this analysis were consistent with those of the primary analysis of PFS (HR = 0.63; 95% CI, 0.46 to 0.86; P = 0.0011) with a median PFS of 7.03 months (95% CI, 4.97 to 9.37) for mogamulizumab and 3.13 months (95% CI, 2.87 to 4.30) for vorinostat.²⁷

Additional (Subgroup) Analyses of PFS

Subgroup analyses of PFS during the randomized treatment period were performed based on investigator assessment for the ITT set and efficacy evaluable set. Results of subgroup analyses for PFS in subgroups of interest identified in Table 6 are summarized in Table 23.

In the subgroup of patients with SS, mogamulizumab was associated with a longer PFS compared with vorinostat (HR = 0.32; 95% CI, 0.21 to 0.49), with a median PFS of 13.30 months for mogamulizumab versus 3.13 months for vorinostat. In patients with stage III or IV disease, mogamulizumab was also associated with a longer PFS over vorinostat (HR = 0.36; 95% CI, 0.26 to 0.51), with a median PFS of 10.90 months for mogamulizumab versus 3.00 months for vorinostat.²⁷

Three post hoc analyses were identified evaluating PFS in patients with advanced-stage MF or SS²⁸ by blood involvement²⁹ and by number and type of prior therapy received.⁴⁶ In the post hoc analysis of patients with advanced-stage MF or SS (stage IIB, IIIA, IIIB, IVA₁, IVA₂, or IVB; n = 150 in the mogamulizumab group and n = 137 in the vorinostat group), the median PFS was higher than the primary analysis but consistent with the subgroup analysis of patients with stage III or IV disease, with a median PFS of 9.40 months for mogamulizumab and 3.07 months for vorinostat.²⁸

PFS by baseline blood classification (B₀, B₁, or B₂) is summarized in Figure 10 in Appendix 3. The results of this post hoc analysis demonstrated that the PFS achieved with mogamulizumab increased with greater blood involvement. After a median efficacy follow-up of 17.0 months, median PFS was higher for mogamulizumab compared with vorinostat in patients with B₂ (11.17 versus 3.30 months) and B₁ (8.63 versus 2.53 months) blood involvement. Median PFS did not differ in patients with B₀ blood involvement (4.70 months versus 4.37 months).²⁹

Results of another post hoc analysis of PFS in patients in the mogamulizumab group of the MAVORIC trial by number and type of prior treatment received is summarized in Table 46, Table 47, and Figure 11 of Appendix 3. The results for PFS for mogamulizumab were consistent with the primary analysis regardless of the number of prior therapies received, ranging from 5.67 months to 10.30 months, or, based on therapeutic class received, ranging from 5.1 months to 10.3 months. Additionally, PFS was consistent with the primary analysis regardless of the specific systemic therapy received, ranging from 5.1 months for romidepsin to 11.2 months for interferon (Figure 11).⁴⁶

PFS by Independent Review

PFS assessed by independent review was a secondary end point of the MAVORIC study. As of the data cut-off date, a total of 232 PFS events had been observed;110 (59.1%) in the mogamulizumab group and 122 (65.6%) in the vorinostat group. Results for PFS based on independent review also favoured mogamulizumab, with an HR of 0.64 (95% CI, 0.49 to 0.84). The median PFS by independent review was similar to the primary analysis, with a median PFS of 6.70 months (95% CI, 5.63 to 9.37) in the mogamulizumab group and 3.83 months (95% CI, 3.00 to 4.70) in the vorinostat group (P = 0.0007).²⁷

Health-Related Quality of Life

HRQoL, as measured by the Skindex-29, FACT-G, and EQ-5D-3L up to cycle 5, was a key secondary end point of the MAVORIC trial.²⁷ At baseline, 173 to 183 patients in the mogamulizumab group and 179 to 185 patients in the vorinostat group completed the HRQoL measures. Completion rates for the HRQoL measures were reduced at cycle 5, with 100 to 101 patients completing HRQoL measures in the mogamulizumab group, and 60 to 61 patients completing HRQoL measures in the vorinostat group.³⁰

Skindex-29

The mean domain scores in the cross-sectional population and the change from baseline in the longitudinal-period population for the Skindex-29 are summarized in Table 24. Symptom scores were the highest at baseline for all domains in both the mogamulizumab and vorinostat arms, with higher scores indicating a higher impact of skin disease. The mean symptom domain scores at baseline for the emotional, functional, and symptom domains were 53.1 (SD = 24.47), 46.1 (SD = 25.50), and 61.2 (SD = 20.35) for mogamulizumab, and 46.3 (SD = 24.24), 40.5 (SD = 24.43), and 58.0 (SD = 19.80) for vorinostat. Overall, both the mogamulizumab and vorinostat groups reported an improvement in Skindex-29 scores for emotions, symptoms, and functioning from baseline to cycle 5 (Table 24).³⁰

In general, the magnitude of the improvement from baseline to cycle 5 was greater with mogamulizumab compared with vorinostat for all domains and at all time points. For example, the greatest mean change from baseline in both the mogamulizumab and vorinostat groups at cycle 5 was in the symptom score domain, with mean changes of −18.0 (SD = 24.19), and −8.2 (SD = 20.73), respectively; the mean difference versus vorinostat was −9.8 (95% CI, −17.13 to −2.56) and favoured mogamulizumab (P = 0.0085).³⁰

Functional Assessment of Cancer Therapy–General

Results for the FACT-G scores by domain and change from baseline are summarized in Table 24. In general, the mean FACT-G total score improved from baseline to cycle 5 for both the mogamulizumab and vorinostat groups. The score in the mogamulizumab group increased from 70.9 (SD = 16.87) to 76.8 (SD = 17.41); the score in the vorinostat group increased from 73.9 (SD = 16.94) to 74.1 (SD = 5.81).³⁰

For the mogamulizumab group, the greatest improvement from baseline compared with vorinostat was seen in the physical well-being domain at cycle 1 (mean score of 0.8 versus −1.5), cycle 3 (mean score of 1.7 versus −2.4), and cycle 5 (2.1 versus −1.9). Similar results were shown for the emotional well-being and functional well-being domains; however, no differences were observed for the social well-being domain, with scores decreasing from baseline to cycle 5, indicating a worse QoL in this domain.³⁰

3-Level EQ-5D

Results for the EQ-5D-3L at each time point and the change from baseline are summarized in Table 24. Results for the individual domains of overall mobility, self-care, usual activities, pain or discomfort, and anxiety were not reported. Based on the index score, HRQoL generally improved from baseline at each time point for both mogamulizumab and vorinostat.³⁰

Mean VAS scores at baseline were 64.7 (SD = 21.53) for mogamulizumab and 60.8 (SD = 20.02) for vorinostat. Mean scores for the VAS were higher at cycle 5 compared with baseline for both mogamulizumab (69.0; SD = 20.30) and vorinostat (63.5; SD = 20.08) groups, with the mean difference versus vorinostat favouring mogamulizumab in cycles 3 and 5.³⁰

Pruritis Evaluation: ItchyQoL and Pruritis Likert Scale

Changes in pruritis via the Pruritis Likert Scale and ItchyQoL was a secondary outcome of the MAVORIC trial. Minimal differences between treatment groups were observed for the Pruritus ItchyQoL emotion, functional, and symptom scores with mean scores from baseline to cycle 5 decreasing for both mogamulizumab, and vorinostat groups.^27,30

Overall Survival

OS was an exploratory end point of the MAVORIC trial and is summarized in Table 26 and Figure 3. At the time of data cut-off, 87 patients (23.4%) had died; 40 in the mogamulizumab group and 47 in the vorinostat group. There was no difference in survival advantage between mogamulizumab and vorinostat (HR = 0.93; 95% CI, 0.61 to 1.43). The median OS was NE for mogamulizumab and was 43.93 months (95% CI, 43.57 to NE) for vorinostat (P = 0.9439).²⁷

Table 26: Summary of Overall Survival (ITT)^a

CI = confidence interval; HR = hazard ratio; ITT = intention-to-treat population; NE = not estimable; OS = overall survival.

^aDoes not include crossover.

^bThe 95% CIs were obtained from the SAS PROC LIFETEST using a log-log transformation.

^cHRs and 95% CIs are based on Cox proportional hazards model with treatment, disease type, disease stage, and region as covariates. P value (2-sided) was obtained from a stratified log-rank test with disease type, disease stage, and region as stratification factors.

^dKaplan-Meier estimate.

Source: MAVORIC (0761 to 010) Clinical Study Report.²⁷

Figure 3: Kaplan-Meier Curve of OS (ITT)

ITT = intention-to-treat population; KW = KW-0761 (mogamulizumab); OS = overall survival; VOR = vorinostat.

Source: MAVORIC (0761 to 010) Clinical Study Report.²⁷

Additional analyses of OS adjusting for the 1-way crossover design using the RPSFT and IPCW methods are summarized in Table 48 of Appendix 3. The results of analyses adjusting for crossover demonstrated no difference in survival advantage for patients treated with mogamulizumab compared with vorinostat, with an HR of 0.739 (95% CI, 0.479 to 1.141) for the RPSFT method and an HR of 0.508 (95% CI, 0.172 to 2.905) for the IPCW method.³⁵ When patients were censored at the time crossover occurred, the HR was 0.709 (95% CI, 0.405 to 1.242).

Time-to-Treatment Failure

TTF was an exploratory end point of the MAVORIC trial, defined as the time from the day of randomization until the discontinuation of randomized treatment due to any reason. The HR for TTF was 0.58 (95% CI, 0.47 to 0.72) in favour of mogamulizumab, with a median TTF of 5.80 months for mogamulizumab compared with 2.87 months for vorinostat.²⁷ The reason for treatment failure is summarized in Table 15. No information on the proportion of patients undergoing a subsequent stem-cell transplant was provided.

Harms

Only those harms identified in the review protocol (Table 6) are reported subsequently.

Adverse Events

TEAEs occurring in the randomized portion (i.e., before crossover) of the MAVORIC trial are summarized in Table 27. A total of 179 (97.3%) and 185 (99.5%) patients had TEAEs with mogamulizumab and vorinostat, respectively. Gastrointestinal disorders were reported more frequently with vorinostat compared with mogamulizumab (93 versus 152 patients; 50.5% versus 81.7%). The most frequently reported TEAEs with mogamulizumab were IRRs (61; 33.2%), drug eruptions (44; 23.9%), diarrhea (43; 23.4%), and fatigue (43; 23.4%). IRRs and drug eruptions occurred at a notably higher rate with mogamulizumab versus vorinostat: there were 61 (33.2%) IRRs for mogamulizumab versus 1 (0.5%) for vorinostat while, for drug eruptions, there were 44 (23.9%) with mogamulizumab versus 1 (0.5%) with vorinostat. The most frequently reported TEAEs with vorinostat were diarrhea (115; 61.8%), nausea (79; 42.5%), fatigue (70; 37.6%), and thrombocytopenia (57; 30.6%).²⁷

Overall, the incidence of specific TEAEs varied by SOC, with TEAEs of infections and infestations, skin and subcutaneous tissue disorders, musculoskeletal and connective tissue disorders, injury, poisoning, and procedural complications reported more frequently with mogamulizumab compared with vorinostat, and TEAEs associated with gastrointestinal disorders, general disorders or administration site conditions, nervous system disorders, investigations, metabolism and nutrition disorders, and blood and lymphatic disorders were reported more frequently in the vorinostat group compared with mogamulizumab, with variation in most of the TEAE preferred terms.²⁷

The overall incidence of grade 3 or greater 3 TEAEs was similar between mogamulizumab (78; 42.4%) and vorinostat (85; 45.7%) and was generally similar across SOCs. However, the incidence of grade 3 or greater infections and infestations was higher with mogamulizumab (32; 17.4%) compared with vorinostat (19; 10.2%), while grade 3 or greater gastrointestinal disorders, general disorders and administration site conditions, and blood and lymphatic system disorders occurred more frequently with vorinostat when compared with mogamulizumab: 4 (2.2%) versus 17 (9.1%), 8 (4.3%) versus 17 (9.1%), and 3 (1.6%) versus 18 (9.7%), respectively.²⁷

In patients who crossed over to mogamulizumab (n = 136), 127 (93.4%) experienced TEAEs of any grade; 47 (34.6%) of these TEAEs were grade 3 or greater.²⁷