FDA Approval Summary: Pembrolizumab plus Lenvatinib for Endometrial Carcinoma, a Collaborative International Review under Project Orbis

In an intensifying effort to support patients with an unmet medical need, the FDA has developed several expedited pathways to make drugs available to patients as rapidly as possible, such as priority review, fast track, breakthrough therapy designation, and accelerated approval (1). FDA has used these expedited programs to effectively streamline the review process for serious and life-threatening oncologic diseases. Additional initiatives through the Oncology Center of Excellence (OCE) include the Real Time Oncology Review (RTOR) where a company may submit datasets for FDA analysis prior to formal application submission allowing FDA to begin their review earlier, and the Assessment Aid (AAid) whereby the company can fill in a template with key information that will focus the FDA review on critical thinking and increase review efficiency (2, 3). Project Orbis in another novel OCE initiative that creates a framework for concurrent submission and review of oncology drugs among international partners.

There is a variable delay, from months to years, between submission of a marketing application in the United States and other jurisdictions, including Australia and Canada. This delay has been captured in several studies that show that companies routinely submit their marketing application first to the FDA, and then, followed by a submission gap of approximately 101–140 days, to Health Canada (HC) and the Australian Therapeutic Goods Administration (TGA; refs. 4, 5). In addition to limiting the anticancer therapies available to patients in these countries, these delays could result in dissimilar standards of care in different geographic areas, and, because of the increasingly global conduct of pivotal clinical trials in cancer drug development, can limit the selection of control arms in future clinical trials. Recognizing these challenges, OCE initiated Project Orbis to allow for concurrent submission and review of oncology products to facilitate earlier access to new anticancer therapies globally.

Prior to embarking on the collaborative review under Project Orbis, FDA selected an upcoming review application, pembrolizumab plus lenvatinib (P+L) for patients with endometrial carcinoma without microsatellite instability-high (MSI-H) or mismatch repair deficiency (dMMR). The combination had been granted breakthrough designation for this population by the FDA in 2018 (6). FDA sought agreement with the drug companies regarding voluntary participation in Project Orbis. Upon acceptance from the drug companies, the process and the application were discussed with HC and TGA, two agencies covered under an international confidentiality agreement with FDA, to ensure that all regulatory agencies were aligned with the application's major features and suitability for review under the differing national regulatory frameworks. In this article, we summarize the FDA, TGA, and HC collaborative review of P+L under Project Orbis for the treatment of patients with advanced endometrial carcinoma that is not MSI-H/dMMR.

Endometrial carcinoma is the sixth most commonly diagnosed cancer and a leading cause of cancer-related deaths in women worldwide (7). Approximately 75% of endometrial carcinoma is diagnosed at an early stage and is typically curable with total hysterectomy and bilateral salpingo-oophorectomy (8). Women with advanced endometrial carcinoma are generally treated with surgery and radiation in combination with standard first-line systemic therapy: carboplatin plus paclitaxel. For patients with recurrent disease following local therapy, or for metastatic endometrial carcinoma, first-line systemic chemotherapy with carboplatin and paclitaxel may provide response rates of about 40%–62%, with expected overall survival of around 13–29 months (9).

Genomic and transcriptomic analyses of endometrial carcinoma have shown that 25%–30% of tumors present with a high frequency of somatic mutations that are attributable to deficiencies in DNA mismatch repair (dMMR), which results in chromosomal changes (expansion or reduction in the length of repetitive sequences in tumor DNA compared with normal DNA) referred to as MSI-H (10, 11). This alteration in chromosomal biology and resultant high mutation rate is thought to result in the increased expression of tumor-associated neoantigens, making MSI-H tumors logical targets for the application of immunotherapy.

Patients who have disease progression or recurrence after initial systemic therapy have limited treatment options. In May 2017, pembrolizumab as a single agent received accelerated approval for refractory MSI-H solid tumors, including endometrial carcinoma, in which it demonstrated a 36% objective response rate (ORR) and durations of response ongoing with a range of 4–17 months (12). However, for patients with tumors that are not MSI-H or dMMR, single-agent cytotoxic chemotherapy is used on the basis of results from small, nonrandomized studies with response rates ranging from 4% to 27% accompanied by short durations of response (13–19). This patient population represents an unmet medical need due to lack of effective treatment options.

The safety and efficacy of the combination of P+L was demonstrated in Study E7080-A001-111/KEYNOTE-146 (hereafter referred to as Study 111), an ongoing, open-label, phase Ib/II, single-arm trial of pembrolizumab in combination with lenvatinib in patients with select metastatic tumor types, including endometrial carcinoma, non–small cell lung cancer, renal cell carcinoma, urothelial cancer, squamous cell carcinoma of the head and neck, and melanoma. The major efficacy outcome measure of the endometrial carcinoma cohort was confirmed ORR and duration of response (DOR) per RECIST 1.1 by independent imaging review (IIR) of pembrolizumab 200 mg i.v. every 3 weeks plus lenvatinib 20 mg orally daily in women with advanced endometrial carcinoma that had progressed following prior systemic therapy and whose tumors were not MSI-H/dMMR (12, 20).

A total of 124 patients were enrolled in the endometrial carcinoma cohort of Study 111. Of these, 108 had previously treated endometrial carcinoma and 94 also had tumors which were not MSI-H/dMMR. This population was the focus of the regulatory authorities' effectiveness and safety review. In this group, the trial demonstrated a confirmed ORR of 38.3% [95% confidence interval (CI), 28.5–48.9] including 10.6% of patients with a complete response (CR) and 27.7% with a partial response (PR). At the time of the data cutoff (January 10, 2019), the median follow-up time was 18.7 months and the median DOR was not reached (range: 1.2+ to 33.1+ months). A total of 25 patients (69% of the 36 responders) had response duration of at least 6 months (see Table 1; refs. 12, 20).

As Study 111 was a single-arm trial, no statistical inferential procedures could be used to evaluate the results. Instead, the effectiveness evaluation relied on the demonstration of improvement over available therapy based on the magnitude of the response rate and an adequate DOR. Because of the single-arm design of Study 111, the treatment effect of each component in the combination could not be isolated. Therefore, monotherapy data were collected from three previously conducted clinical trials (Study204 for lenvatinib monotherapy and KEYNOTE-158 and KEYNOTE-028 for pembrolizumab monotherapy) and provided supportive information for estimating the isolation of treatment effect (Table 2; refs. 21, 22). While exploratory post hoc cross-trial comparisons indicated a numerically higher ORR with the combination therapy over the individual treatments, FDA also conducted exploratory post hoc analyses using propensity score approach to evaluate the contribution of each component using data from these external monotherapy trials (23). Propensity scores were estimated using a logistic regression model, in which treatment status was regressed on observed baseline characteristics including demographic and baseline disease characteristics. The baseline factors chosen were based on what was collected and accessible in the compared trials. However, this approach cannot guarantee that all measured and unmeasured baseline factors will be balanced. The results of the exploratory propensity score analyses are consistent with those seen in an unadjusted comparison of the combination treatment versus each individual treatment in this population.

The safety analysis included Study 111 and the monotherapy trials to evaluate the contribution of each drug to the safety profile of the P+L combination. Fatal adverse reactions (AR) occurred in 3% of patients receiving P+L, including gastrointestinal perforation, reversible posterior leukoencephalopathy syndrome with intraventricular hemorrhage, and intracranial hemorrhage (12, 20). Permanent discontinuation due to AR occurred in 21% of patients and serious adverse events occurred in 52% of patients receiving P+L combination. Lenvatinib alone was dose modified (interrupted and/or reduced) in 88% of the patients and pembrolizumab alone was dose interrupted in 49%. As expected with the combination regimen, the rate of AR, and dose modifications was higher in Study 111 compared with the monotherapy studies, however, all the ARs were consistent with the known safety profiles of pembrolizumab and lenvatinib. Provider awareness and dose modifications allowed the majority of patients to remain on treatment (12, 20).

Key ARs of interest with lenvatinib, identified as clinically significant events (CSE, which include ARs that have been identified in clinical trials across the lenvatinib development program), occurred in 86 (91.5%) patients. Adverse events (AE) of special interest (AESI: immune-mediated AEs or nonimmune-mediated events of infusion-related reactions) with pembrolizumab occurred in 54 (57.4%) of patients. Both CSEs and AESIs occurred at higher frequencies in Study 111 compared with monotherapy studies, Study 204 and KEYNOTE-158, respectively.

As clinical trials are increasingly being conducted internationally, availability of newer drugs globally will provide greater uniformity in standards of treatment, which may lead to more efficient drug development and earlier access to therapies for people with cancer. Programs such as Project Orbis establish a framework for concurrent submission and review of oncology therapies by international partners. The FDA, the Australian TGA, and HC collaborated for the first review under this pilot, which in this case allowed for near-simultaneous approval in all three countries (24).

During this collaborative approach to the applications, FDA undertook its standard review approach (review elements were not divided among the agencies), examining safety and efficacy data at the level of the primary source data, while TGA and HC each performed their standard evaluation. In addition, this application was reviewed under the OCE's RTOR pilot program and the AAid pilot. By participating in RTOR, the companies submitted datasets for FDA analysis upon completion of the clinical trial, but before they had completed all of the summary application submission documents, allowing FDA to undertake detailed data analyses without having to wait for the formal submission of the entirety of the dossier (2). The AAid provided a dynamic platform for engagement among the review disciplines from the three agencies, where key regulatory questions and analyses were discussed during the analytic phase of the review (3). Findings by each of the participating agencies were discussed in a series of teleconferences, which also served as an opportunity for discussion about differences in review strategies and regulatory frameworks (Fig. 1). Through this mechanism, the teams were able to discuss data supporting the safety and effectiveness of the combination of P+L in this population of women with metastatic endometrial carcinoma.

This approval action represents the first approved therapy for patients with endometrial carcinoma that has progressed following prior systemic therapy whose tumors are not MSI-H or dMMR, across various histologies. Uncertainties regarding the benefit of the P+L combination stem from the single-arm trial design and contribution of effect, however, the magnitude of ORR improvement seen with the combination was greater than that of the individual components and exploratory analyses supported this finding. Two phase III randomized clinical trials for the P+L combination were ongoing at the time of accelerated approval: Study E7080-G000-309/KEYNOTE-775, comparing the efficacy and safety of P+L versus treatment of physician's choice in patients with advanced endometrial carcinoma, who have received one prior platinum-based chemotherapy regimen; and Study E7080-G000-313/MK-7902-001, comparing the efficacy and safety of P+L versus chemotherapy for the first-line treatment of patients with advanced or recurrent endometrial carcinoma. These ongoing trials strengthened the application by ensuring that data necessary to verify clinical benefit will be obtained. As enrollment of both trials was ongoing at the time of approval, and the first-line trial is in a different treatment setting, accrual is not excepted to be of concern.

Labeling was negotiated separately between each regulatory agency and the companies and varies slightly between the agencies, based on the national healthcare setting and regulatory framework. The clinical trial submitted as evidence for approval enrolled only women with metastatic endometrial carcinoma that had progressed following platinum-based chemotherapy; FDA determined that there is no scientific rationale that other women with incurable locally advanced endometrial carcinoma that is not curable with surgery or radiation that had progressed following treatment would not also benefit to a similar extent from this combination. Similarly, patients who progressed following platinum-based systemic therapy and those who progressed following nonplatinum regimens should also similarly respond to this therapy. Thus, the indication approved by the FDA and TGA is applicable to a slightly broader population of women than that included in the Study 111, while the HC indication is more closely aligned to the patient population in the Study 111 (12, 20, 25–28). Throughout the process, drug labels were exchanged between the different regulatory agencies to learn about and discuss any potential differences.

Participating in Project Orbis did not alter FDA's overall approach to the review, the evaluation of the risk–benefit assessment, or delay the approval of the P+L. FDA is working on expanding partnerships with other regulatory agencies and continues to hold monthly teleconferences with TGA, HC, the European Medicines Agency, Japan's Pharmaceuticals and Medical Devices Agency, and Switzerland's Swissmedic. In addition, FDA and China's National Medical Products Administration have initiated a quarterly meeting to discuss nonproduct-specific regulatory issues facing worldwide drug development. As this pilot project continues, FDA will offer drug companies an opportunity to participate in Project Orbis based on the unmet medical need and/or topline efficacy and safety results of an application. Companies may also request consideration for participation in Project Orbis. While this application was a supplemental application, future efforts will also focus on how new molecular entities could be included in Project Orbis.

P+L for the treatment of women with advanced endometrial carcinoma that is not MSI-H/dMMR and that has progressed following systemic therapy has a favorable benefit–risk profile and supports accelerated approval (Table 3). Continued approval for this indication may be contingent upon verification and description of clinical benefit in another trial. Post-approval randomized confirmatory studies are ongoing, results of which will enhance our understanding of the effectiveness of this combination.

This was the first evaluation facilitated through Project Orbis, with a collaborative review approach taken between FDA, TGA, and HC, and completed nearly 3 months ahead of the FDA priority review deadline (Fig. 1). Participating in programs such as Project Orbis enables learning and consistency among regulators, strengthens international relationships, and potentially allows earlier access to important new treatments for the global community. At the same time, collaborative reviews highlight similarities and differences between regulations in the different countries. FDA, TGA, and HC plan to continue building these partnerships, expanding to include other regulatory agencies including Swissmedic and Singapore's Health Science Authority in the future.

No potential conflicts of interest were disclosed.

The Editor handling the peer review and decision-making process for this article has no relevant employment associations to disclose.