FDA Decision Alerts August 14, 2020

CDER-Approved NDA for TAZVERIK™ (tazemetostat)

Christina Purpura and Mandy Patrick
Contributing writers, Aetion

On January 23, 2020, the FDA approved Epizyme’s TAZVERIK™ (tazemetostat) for patients “16 years and older with metastatic or locally advanced epithelioid sarcoma not eligible for complete resection.” Key findings from the FDA’s Multi-Discipline Review, Administrative and Correspondence Documents, and the Sponsor Briefing Document for NDA 211723:

Safety and efficacy for TAZVERIK are based on an open label, single-arm cohort (Cohort 5) of a multi-center study, EZH-202 (NCT02601950, n=62). Cohort 5 demonstrated an objective response rate (ORR) of 15% (95% CI: 7, 26); according to the Multi-Discipline Review, ORR is defined as “confirmed complete response (CR) or partial response (PR) from the start of treatment until disease progression or the start of subsequent anti-cancer therapy as per RECIST 1.1 criteria.” The sponsor also submitted a natural history study, Study EZH-1001 (NCT03837678; n=69) among patients with epitheliloid sarcoma (ES) who had not received tazemetostat.

Intent of the RWE study
The applicant shared a proposal with the FDA to submit an ES natural history study as a control arm to support full approval. The specific study goals were to determine safety and efficacy in the real world for patients with ES who require systemic therapy, and to improve understanding of the natural history of ES. However, the agency provided feedback during an end-of-phase 2 meeting on March 3, 2017: It did not agree that the proposed study could provide a control arm to Cohort 5 of EZH-202 to support regular approval. Instead, the agency stated that the natural study could “provide a better understanding of the natural history and outcomes of patients with ES.”

In the same document, the FDA later writes “while these comments are intended to enhance the interpretability of the data from the Natural History study, the FDA considers [real-world overall response rate (rwORR)] not comparable to ORR as assessed on a clinical trial, and considers cross-trial comparisons of time-to-event endpoints not valid. It is thus unlikely that a response to these comments will result in FDA agreement that the ES Natural History Study can be used as a ‘control arm’ for the purposes of regular approval.”

Protocol for RWE generation
Study EZH-1001, a multi-center, non-interventional retrospective medical chart review, involved ES patients who were initiated on systemic chemotherapy between 2000 and 2017. Data were abstracted from medical charts at five U.S. academic cancer centers; specifically:

  • demographic characteristics;
  • disease history;
  • prior treatments for ES; and
  • tumor response information, which were then categorized and referred to as rwORR.

rwORR rate was the primary endpoint, which was “defined as the proportion of patients who have a documented radiological scan showing clinician assessed complete response or less-than-complete response, of any duration, defined for each regimen and by line.” Real-world duration of response (DOR) and overall survival (OS) were secondary endpoints.

Data were analyzed by type of treatment(s) received, as well as by line of therapy.

In addition to Study EZH-1001, the applicant conducted an extensive literature review on the effectiveness of approved therapies for ES patients. However, available data were limited (e.g., small patient numbers, retrospective studies), and there were differences in response criteria, eligibility, and patient populations. Therefore, the reviewer could “only conclude that response rates for patients with epithelioid sarcoma treated with tazemetostat or approved therapies do not appear different from those of patients with other forms of [soft tissue sarcoma].”

Outcome of the RWE submission
For patients treated with anthracycline-based regimen, gemcitabine-based regimen, pazopanib, or other available therapies, Study EZH-1001 demonstrated low to modest overall response rate, short durability of responses, and limited OS.

The FDA critiqued the study protocol because it did not “provide adequate detail regarding quality of data, validity of endpoint assessments, and design choices, rendering the results of the study uninterpretable. For general principles regarding observational studies, please refer to ‘Guidance for Industry and FDA Staff: Best Practices for Conducting and Reporting Pharmacoepidemiologic Safety Studies Using Electronic Healthcare Data.’” On design choices, the FDA continues, “an observational study whose intent is to serve as an historical control for single arm data should be designed such that the patient populations to be compared in the analyses are as similar as possible.” Differences in eligibility criteria:

  • age;
  • years studied;
  • INI-1 status;
  • cancer therapy(ies) prior to enrollment; and
  • the minimum length of the pre-index period for assessing patient characteristics.

In addition, Study EZH-1001 did not specify methods to evaluate or control for confounding factors in a comparison of outcomes between EZH-202 and EZH-1001.

Supplemental NDA, for follicular lymphoma
In addition to the first NDA, the FDA granted accelerated approval for TAZVERIK’s second NDA, NDA 213400, on June 18 for adult patients with relapsed or refractory follicular lymphoma. The approval was based on two open-label, single-arm cohorts (Cohorts 4-5) of a multi-center trial, Study E7438-G000-101 (NCT01897571). While the Multi-Discipline Review is not available, to fully characterize the safety and efficacy of TAZVERIK, the applicant needs to submit a final report to the FDA in 2026 containing data from clinical trials, postmarket reports, compassionate use, RWE, and other sources.


eBook: 2021 update

The Role of Real-World Evidence in FDA Approvals

Download here