FDA Accelerated Approval of Pembrolizumab for Recurrent Locally Advanced or Metastatic Merkel Cell Carcinoma

Abstract

On December 19, 2018, the Food and Drug Administration granted accelerated approval to pembrolizumab (KEYTRUDA, Merck & Co. Inc., Whitehouse Station, NJ) for adult and pediatric patients with recurrent locally advanced or metastatic Merkel cell carcinoma (MCC). Approval was based on Cancer Immunotherapy Trials Network protocol 9, also known as KEYNOTE‐017 (NCT02267603), a multicenter, nonrandomized, open‐label trial that enrolled 50 patients with recurrent locally advanced or metastatic MCC who had not received prior systemic therapy for their advanced disease. The major efficacy outcome measures were overall response rate (ORR) and response duration assessed by blinded independent central review per RECIST 1.1. The ORR was 56% (95% confidence interval: 41, 70) with a complete response rate of 24%. The median response duration was not reached. Among the 28 patients with responses, 96% had response durations of greater than 6 months and 54% had response durations of greater than 12 months. The most common adverse reactions of pembrolizumab reported in at least 20% of patients who received pembrolizumab as a single agent were fatigue, musculoskeletal pain, decreased appetite, pruritus, diarrhea, nausea, rash, pyrexia, cough, dyspnea, constipation, pain, and abdominal pain.

Implications for Practice

This report presents key information on the basis for the Food and Drug Administration's accelerated approval of pembrolizumab for the treatment of recurrent locally advanced or metastatic Merkel cell carcinoma, including efficacy and safety information. This approval provides patients and physicians with an additional treatment option for this aggressive and life‐threatening carcinoma.

Short abstract

This report summarizes key information on the basis for the accelerated approval of pembrolizumab for the treatment of recurrent locally advanced or metastatic Merkel cell carcinoma.

Introduction

Merkel cell carcinoma (MCC) is a rare, aggressive neuroendocrine carcinoma of the skin, characterized by its expression of cytokeratin 20 and an association with the Merkel cell polyomavirus (MCPyV) in the majority (approximately 80%) of patients 1, 2, 3. An estimated 2,500 new patients are diagnosed each year in the U.S. with MCC, with increasing annual incidence rates 4. Risk factors for this disease include ultraviolet radiation exposure, infection with the Merkel cell polyomavirus, age, and immunosuppression. The mean age of presentation is approximately 75 years 5. More than 30% of patients will develop distant metastatic disease 6. The 5‐year and 10‐year relative survival rates for patients with metastatic MCC are reported as 25% and 21%, respectively, according to analyses of Surveillance, Epidemiology, and End Results data 7, 8.

MCC generally presents as a firm, painless, rapidly enlarging, dome‐shaped cutaneous nodule located in a sun‐exposed area such as the head, neck, or upper extremities 4. Wide local excision with at least a 2‐ to 3‐cm minimum margin is the standard of care for localized MCC. Depending on the tumor location and the excision margins, adjuvant radiation may be administered following resection. For patients with disease present in local lymph nodes, the approach may include a wider dissection with lymphadenectomy and adjuvant radiation 5, 9. Prior to the December 2018 accelerated approval of pembrolizumab for patients with recurrent locally advanced or metastatic MCC, the topic of this report, avelumab was the only Food and Drug Administration (FDA)‐approved drug for this disease. Avelumab was approved under FDA accelerated approval regulations and had not been granted regular approval prior to the approval of pembrolizumab for MCC.

Prior to the approval of avelumab and pembrolizumab, patients with distant metastatic disease would receive off‐label cytotoxic chemotherapy tailored to the individual patient's age and baseline comorbidities. The common chemotherapy drugs used were cisplatin or carboplatin in combination with etoposide, single‐agent topotecan, and cyclophosphamide, doxorubicin, and vincristine. The reported response rates for patients with newly diagnosed metastatic MCC were between 50% and 60%; however, responses to chemotherapy were of limited duration (median 85 days) and chemotherapy does not improve the overall survival (OS) of patients with metastatic MCC 10, 11, 12.

Pembrolizumab is a monoclonal antibody that binds to the programmed death receptor‐1 (PD‐1) and blocks its interaction with programmed death ligand‐1 (PD‐L1) and PD‐L2, releasing PD‐1 pathway–mediated inhibition of the immune response, including the antitumor immune response. Pembrolizumab was previously approved for the treatment of multiple solid tumors, including indications for melanoma, non‐small cell lung cancer, small cell lung cancer, head and neck squamous cell cancer, classical Hodgkin lymphoma, primary mediastinal large B‐cell lymphoma, urothelial carcinoma, microsatellite‐high cancer, gastric cancer, cervical cancer, esophageal cancer, endometrial carcinoma, renal cell carcinoma, and hepatocellular carcinoma. A summary of information on pembrolizumab is in Table 1.

Table 1.

Pembrolizumab background information

Structurea	Anti‐PD‐1 monoclonal antibody MK‐3475 humanized mouse monoclonal (228‐L‐proline(H10‐S > P)) gamma 4 heavy chain (134‐218')‐disulfide with humanized mouse monoclonal kappa light chain dimer (226‐226”:229‐229”)‐bisdisulfide immunoglobulin G4, anti‐(human programmed cell death 1)
Mechanism of action	Binds to PD‐1 and blocks its interaction with PD‐L1 and PD‐L2, releasing PD‐1 pathway–mediated inhibition of the immune response, including the antitumor immune response. In syngeneic mouse tumor models, blocking PD‐1 activity resulted in decreased tumor growth.
Pharmacokineticsb	Pharmacokinetics of pembrolizumab were characterized using a population PK analysis with concentration data collected from 2,993 patients with various cancers who received pembrolizumab doses of 1–10 mg/kg every 2 weeks, 2–10 mg/kg every 3 weeks, or 200 mg every 3 weeks. Steady‐state concentrations of pembrolizumab were reached by 16 weeks of repeated dosing with an every‐3‐week regimen and the systemic accumulation was 2.1‐fold. The peak concentration, trough concentration, and area under the plasma concentration versus time curve at steady state of pembrolizumab increased dose proportionally in the dose range of 2–10 mg/kg every 3 weeks.
Prior approvals	Approved for melanoma, non‐small cell lung cancer, small cell lung cancer, head and neck squamous cell cancer, classical Hodgkin lymphoma, primary mediastinal large B‐cell lymphoma, urothelial carcinoma, microsatellite‐high cancer, gastric cancer, cervical cancer, esophageal cancer, hepatocellular carcinoma, renal cell carcinoma, and endometrial carcinoma.

^aNCI Drug Dictionary. Pembrolizumab. Available at https://www.cancer.gov/publications/dictionaries/cancer‐drug/def/pembrolizumab.

^bU.S. Prescribing Information, KEYTRUDA (pembrolizumab). Available at https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/125514s055s056lbl.pdf.

Abbreviations: PD‐1, programmed death receptor‐1; PD‐L1, programmed death ligand‐1; PK, pharmacokinetics.

Trial Design

KEYNOTE‐017 (NCT02267603) was a multicenter, nonrandomized, open‐label, single‐arm clinical trial to evaluate the efficacy and safety of pembrolizumab in patients with recurrent locally advanced or metastatic MCC who had not received prior systemic chemotherapy for advanced MCC. The investigators’ analyses of the data have been published 13, 14. Patients were eligible for the trial if they had biopsy‐proven metastatic MCC or locoregional MCC that had recurred following standard locoregional therapy with surgery or radiation therapy. If administered, prior radiation therapy must have been completed more than 2 weeks prior to the initiation of study treatment. Patient disease had to be measurable per RECIST 1.1 and assessed by computed tomography (CT) imaging, except for skin lesions not measurable by CT scan, in which case measurements with a caliper or flexible ruler were permitted. Patients who had received prior systemic therapy for MCC, or who had participated in a study of an investigational systemic agent to treat MCC, were not eligible. Patients received pembrolizumab 2 mg/kg every 3 weeks intravenously for 2 years, or until progressive disease (PD) or unacceptable toxicity. Imaging assessments were performed at week 13, then every 9 weeks during the first year, and then every 12 weeks during the second year of treatment.

The primary efficacy endpoint of the study for regulatory submission was overall response rate (ORR) assessed by Blinded Independent Committee Review (BICR) per modified RECIST 1.1, defined as the proportion of participants who had a confirmed complete response (CR) or partial response (PR). The ORR was described using a point estimate of response rate and 95% confidence interval (CI) based on the Clopper‐Pearson methods for the exact binomial distribution. The modifications of RECIST for the independent review for this study included a provision to not allow chest x‐ray measurements and minor changes to lymph node criteria.

The secondary endpoints of the study included duration of response (DOR) and estimates of progression‐free survival (PFS) and OS. DOR was defined as the time from the first documented evidence of confirmed CR or PR until disease progression. PFS was defined as the time from allocation in the study to the first documented evidence of PD as per RECIST 1.1 or death due to any cause. OS was defined as the time from allocation to death due to any cause. For these secondary endpoints, the survival curve and median time were estimated using the Kaplan‐Meier method. Patients who were alive at the time of analysis were censored at the database cutoff date for OS analysis, and DOR and PFS were censored at the last assessment date. If a substantial amount of primary endpoint data was missing (at least 1 value missing from more than 20% of patients), analyses of the ORR would be performed using parametric generalized linear models fit by maximum likelihood. A generalized linear model for the ORR would use a binomial error distribution and include all available baseline predictors of the missing outcomes as covariates.

Results

A total of 50 patients with MCC were enrolled and received treatment in KEYNOTE‐017. The demographics of the 50 patients are provided in Table 2. Most patients had distant metastatic disease (86%), whereas a minority had locally advanced disease (stage IIIB, 14%). Eighty‐four percent of patients had prior surgery, and 70% received prior radiation. Eighty‐six percent of patients had no prior medical anticancer therapy; of the remaining patients, 8% received intratumoral therapy and 6% received prior adjuvant therapy. Most patients were classified as PD‐L1 positive (68%), with 28% PD‐L1 negative, and the status of the remaining 4% was unknown.

Table 2.

Demographic characteristics of all enrolled patients

Demographic characteristics	Pembrolizumab (n = 50), n (%)
Age, median, years	70.5
Age <65 years	10 (20.0)
Age ≥65 years	40 (80.0)
Sex
Male	34 (68.0)
Female	16 (32.0)
Race
White	45 (90.0)
Asian	1 (2.0)
Unknown	4 (8.0)
ECOG
0	24 (48.0)
1	26 (52.0)

Abbreviation: ECOG, Eastern Cooperative Oncology Group.

Efficacy

Based on the BICR assessment, among the 50 patients in the All‐Subjects‐As‐Treated population, 28 experienced a confirmed objective response (CR or PR), resulting in an ORR of 56% (95% CI: 41, 70). Twelve patients (24%; 95% CI: 13, 38) experienced a CR, and 16 patients (32%; 95% CI: 20, 47) experienced a PR. The results are shown in Table 3. In Table 4, an analysis of ORR based on different characteristics is shown.

Table 3.

Results of response and duration of response

Efficacy parameter	Pembrolizumab 2 mg/kg Q3W (n = 50)
Overall response rate (CR + PR)
Number of responses	28
% (95% CI)a	56 (41, 70)
Complete responses, n (%)	12 (24)
Partial responses, n (%)	16 (32)
Duration of response
Median in monthsb	Not reached
Min, Max	5.9, 34.5+

Based on the exact binomial distribution.

Based on the Kaplan‐Meier method.

Abbreviations: CI, confidence interval; CR, complete response; PR, partial response; Q3W, every 3 weeks.

Table 4.

Subgroup analysis of ORR by age, gender, and ECOG performance status

Characteristics	No. of objective responses (CR + PR)/no. of patients	ORR% (95% CI)a
Overall	28/50	56 (41,70)
Age, years
<65	4/10	40 (12, 74)
≥65	24/40	60 (43,75)
Gender
Male	18/34	53 (35, 70)
Female	10/16	63 (35, 85)
ECOG performance status
0	15/24	63 (41, 81)
1	13/26	50 (30, 70)

Based on the binomial exact confidence interval method.

Abbreviations: CI, confidence interval; CR, complete response; ECOG, Eastern Cooperative Oncology Group; ORR, overall response rate; PR, partial response.

The Kaplan‐Meier estimate of the median duration of response was not reached among the 28 responders with BICR‐assessed confirmed objective response. The range of duration of response was 5.9–34.5 months (ongoing), as shown in Table 3. As of the database cutoff date of February 6, 2018, 20/28 (71%) of the responses were ongoing. Among the responders, the median time to response was 2.8 months with a range of 1.5–9.7 months. Among the 28 responders, the number of patients with an observed duration of response greater than 6 months was 27 (96%), greater than 12 months was 15 (54%), greater than 18 months was 9 (32%), and greater than 24 months was 5 (18%). PFS and OS were assessed as secondary endpoints, but as these time‐to‐event endpoints were not interpretable given the single‐arm trial design, FDA did not consider these endpoints in the assessment of substantial evidence of effectiveness.

Safety

The safety of pembrolizumab was primarily evaluated in 50 patients with locally advanced or metastatic MCC that recurred following locoregional therapy with surgery or radiation therapy. Each patient received at least one dose of pembrolizumab 2 mg/kg intravenously, and the safety monitoring period was from the time the patient signed the informed consent through 30 days after the end of treatment. The median duration of treatment was 6.6 months (range 0.03–23.6 months), with patients receiving a median of 10.5 infusions. The safety review of pembrolizumab was also supported by results from a pooled population (n = 2,799) consisting of patients with various advanced solid tumors, largely melanoma, who received pembrolizumab at a dose of 1–10 mg/kg every 2–3 weeks in KEYNOTE‐001, KEYNOTE‐002, KEYNOTE‐006, or KEYNOTE‐010.

In KEYNOTE‐017, there were no fatal adverse events (AEs) that were clearly pembrolizumab‐related. Forty‐four percent of patients experienced at least one serious AE (SAE). The most frequently reported SAE aside from disease progression was dehydration (4 [8.0%] patients). SAEs of increased aspartate aminotransferase (AST) and embolism were each reported for 3 (6.0%) patients, compared with 6 (0.2%) and 9 (0.3%) patients, respectively, in the Reference Safety Database (RSD). In KEYNOTE‐017, all other SAEs were each reported for ≤2 patients. Eleven patients (22%) experienced an SAE assessed by the applicant as treatment‐related in KEYNOTE‐017, which was higher than observed in the RSD (10%). In CITN‐09/KEYNOTE‐017, the most frequently reported treatment‐related SAEs were increased AST and pneumonitis, each reported in 2 (4.0%) patients, compared with 4 (0.1%) and 44 (1.6%) patients, respectively, in the RSD. All other treatment‐related SAEs, which were largely immune‐mediated, including colitis, rash, myocarditis, or lichenoid keratitis, were reported for 1 participant each.

Pembrolizumab was discontinued in nine patients because of adverse events. The most common reason for discontinuation was pancreatitis in two patients (4%), followed by one event for each of the following: adrenal insufficiency, increased blood alkaline phosphatase, colitis, myocarditis, paraneoplastic syndrome, pneumonitis, polyarthritis, or tuberculosis (2% each). Fourteen patients (28%) experienced 21 adverse events of special interest (AEOSIs) on KEYNOTE‐017. The most commonly reported AEOSIs were hypothyroidism (n = 3, 6%) and pneumonitis (n = 3, 6%). Pancreatitis and maculopapular rash were each reported in two (4%) patients. Of the five patients with grade 3 or 4 AEOSIs, one (2.0%) participant experienced grade 4 myocarditis, one (2.0%) participant experienced grade 3 colitis and pancreatitis, one (2.0%) participant experienced grade 3 pneumonitis, and two (4.0%) patients experienced grade 3 rash maculo‐papular. All grade 3–4 AEOSIs resolved. Five patients had an AEOSI that was not resolved as of the database cutoff date (hypothyroidism in three patients; adrenal insufficiency, hyperthyroidism, and pneumonitis each in one patient). Treatment‐emergent adverse events, which occurred in >20% of patients, included fatigue (58%), anemia (32%), constipation (28%), hyperglycemia (26%), upper respiratory tract infection (26%), decreased appetite (22%), increased alanine aminotransferase (22%), rash, and pruritis (22% each).

Overall, the safety of pembrolizumab is consistent with the expected toxicity profile of immunologically mediated anticancer therapies, and the safety of pembrolizumab for the treatment of recurrent, locally advanced or metastatic MCC is comparable to the safety of pembrolizumab across indications as evaluated through comparison to the RSD.

Pediatrics

KEYNOTE‐017 did not include pediatric patients. The FDA‐approved indication includes pediatric patients based on extrapolation of pharmacokinetics and target exposures from adult to the pediatric population, as established in KEYNOTE‐051 (NCT02332668), which enrolled patients between 6 months and <18 years of age with locally advanced or metastatic solid tumors or lymphoma who had failed prior therapy, had no available standard therapy, or were not eligible for standard therapy. The safety profile in these pediatric patients was similar to that seen in adults and is described in product labeling.

Discussion

Locally advanced or distant metastatic Merkel cell carcinoma is a serious, aggressive, life‐threatening disease with a reported 5‐year survival rate of 25%. There are currently few treatment options, and at the time of the pembrolizumab approval, avelumab was the only FDA‐approved therapy for this disease. The study that supported approval of avelumab for the treatment of patients with metastatic MCC (JAVELIN Merkel 200) was an open‐label, single‐arm, multicenter study conducted in patients with histologically confirmed metastatic MCC whose disease had progressed on or after chemotherapy administered for distant metastatic disease 15. KEYNOTE‐017 was conducted in patients who had not received prior systemic therapy for their advanced disease; thus, any comparison of response rates across trials should be interpreted with caution in light of the different populations studied.

Because off‐label cytotoxic treatment options appear to have short‐term benefits, there is an unmet medical need for patients with metastatic MCC.

The Applicant sought accelerated approval of pembrolizumab for this indication based on a 56% response rate including 24% of patients with a complete response. Response rate (e.g., as determined using RECIST criteria) can be assessed in single‐arm trials because generally tumors do not shrink in the absence of treatment. When determining whether an observed response rate is sufficient to support accelerated or regular approval, FDA considers multiple factors including the magnitude of the effect, durability of the responses (considering the specific cancer), unmet medical need (influenced by the availability of other therapies), depth of responses (e.g., complete responses), burden of underlying disease in the studied population, and the frequency of the cancer.

FDA believes that the response rate of the magnitude and durability observed in KEYNOTE‐017, if confirmed based on additional data (including data on durability), could represent clinical benefit in and of itself in this ultrarare, aggressive cancer. This includes approximately one out of four patients without radiographically (or visually) detectable disease following treatment with pembrolizumab.

Given the results of KEYNOTE‐017, FDA does not believe it would be feasible or appropriate to require patients to be randomized to traditional cytotoxic chemotherapy (or placebo). The results from KEYNOTE‐017 compare favorably with off‐label use of chemotherapy that produces nondurable response rates with considerable toxicity. Furthermore, crossover to anti‐PD(L)‐1 targeted monoclonal antibodies (along with the rarity of the cancer) would likely preclude the ability to power a randomized study to detect an effect on overall survival.

FDA granted accelerated approval to pembrolizumab for this application (Table 5). FDA may grant accelerated approval for a drug that demonstrates an effect on an intermediate endpoint that is reasonably likely to predict clinical benefit or on a clinical endpoint that can be measured earlier than irreversible morbidity or mortality (IMM) that is reasonably likely to predict an effect on IMM or other clinical benefit 16. FDA considered the effect of pembrolizumab on ORR in this setting with promising but incompletely characterized durability of response to be an intermediate clinical endpoint where the effect on ORR and DOR would need to be verified in postapproval confirmatory studies. FDA has taken this approach in other applications, including the site‐agnostic approval of pembrolizumab for the treatment of patients with microsatellite‐instable tumors. A trial to confirm the clinical benefit of pembrolizumab in patients with recurrent locally advanced or metastatic MCC is currently ongoing (NCT03783078).

Table 5.

FDA risk–benefit assessment

Dimension	Evidence and uncertainties	Conclusion and reasons
Analysis of condition	MCC is a rare, aggressive neuroendocrine carcinoma of the skin, characterized by its expression of cytokeratin 20 and an association with MCV. Factors associated with increased risk for developing MCC include ultraviolet radiation exposure, infection with MCV, advanced age, and immunosuppression, although most patients with MCC are immunocompetent. The mean age at presentation is approximately 75 years. The 5‐year survival rate for patients with MCC is 25% for tumors with distant metastases. More than 30% of patients will develop distant metastatic disease.	Metastatic MCC is incurable and represents a disease with only one FDA‐approved therapy; therefore, there is an unmet medical need and a need for additional drugs proved to be effective for the disease.
Current treatment options	Avelumab The most common drugs used off‐label are cisplatin or carboplatin in combination with etoposide, single‐agent topotecan, or cyclophosphamide, doxorubicin, and vincristine. MCC is generally a chemo‐sensitive tumor with response rates between 50% and 60% for patients with newly diagnosed metastatic MCC; however, responses are not durable (median duration of 85 days from reports in the published literature), and chemotherapy has not been shown to improve OS in patients with metastatic disease.	There are few treatment options for patients with metastatic MCC and only one approved drug, avelumab. Avelumab is approved under the accelerated approval regulations. Existing chemotherapy regimens offer limited benefit in terms of duration of responses, when responses are observed.
Benefit	The primary evidence of effectiveness supporting this supplemental Biologics License Application (BLA) is from KEYNOTE‐017, a multicenter, nonrandomized open‐label trial of pembrolizumab in 50 patients. The assessment of efficacy is based on the endpoints of confirmed ORR and DOR as determined by a blinded independent central review. Among the 50 patients in the trial, 28 experienced a confirmed objective response (CR or PR), resulting in an ORR of 56% (95% CI: 41.3, 70.0). Twelve patients (24%; 95% CI: 13.1, 38.2) experienced a CR, and 16 patients (32%; 95% CI: 19.5, 46.7) experienced a PR. The median DOR was not reached, and the range of responses were 5.9 to 34.5 months.	KEYNOTE‐017 was a well‐conducted trial demonstrating a clinically meaningful response rate for a serious and life‐threatening rare disease and with significantly longer response durations observed or reported with chemotherapy. In addition, the data observed for pembrolizumab in the relevant patient population compare favorably to those of the only drug approved for this indication. Because both pembrolizumab and avelumab will have been approved under the accelerated approval regulations, confirmation of the clinical effects for both drugs is pending.
Risk and risk management	In KEYNOTE‐017, there were no fatal AEs that were clearly pembrolizumab‐related. Forty‐four percent of patients experienced at least one SAE, comparable to the pooled reference safety database (37.2%). Fourteen patients (28%) experienced 21 adverse events of special interest, including immune‐mediated events as well as infusion‐related reactions. The frequency of these adverse events was like that of the pooled safety population (21.2%). Nine patients (18%) discontinued pembrolizumab because of adverse events. There is a favorable benefit–risk assessment for pembrolizumab for the treatment of patients with recurrent locally advanced or metastatic MCC.	The observed safety profile is acceptable when assessed in the context of the treatment of a life‐threatening disease. The incidence and severity of immune‐mediated adverse reactions in KEYNOTE‐017 is like that reported in the pooled safety database for pembrolizumab administered as a single agent. Significant and serious adverse reactions, including immune‐mediated adverse reactions, are addressed in the Warnings and Precautions section and the dose modification recommendations included in product labeling. There were no significant safety concerns identified during the review of the application requiring risk management beyond labeling or warranting consideration for a Risk Evaluation and Mitigation Strategy.

Abbreviations: AE, adverse event; CI, confidence interval; CR, complete response; DOR, duration of response; FDA, Food and Drug Administration; MCC, Merkel cell carcinoma; MCV, Merkel cell polyomavirus; ORR, objective response rate; PR, partial response; SAE, serious adverse event.

Conclusion

The availability of anti‐PD‐(L)1 therapy, with approvals for both avelumab and pembrolizumab, has fundamentally changed the landscape of therapy for patients with Merkel cell tumors. As reported by Nghiem and colleagues, response rate to pembrolizumab appeared similar in patients with both MCPyV‐positive and MCPyV‐negative tumors 13. Future studies may determine the effects of anti‐PD‐(L)1 therapy in the adjuvant setting (e.g., NCT03712605; NCT03271372; NCT02196961). In the metastatic setting, it is expected that trials will assess the effects of drugs (or radiation) in combination with checkpoint inhibitors in the first‐line setting or to assess the effects of investigational drugs in patients who have not responded to or progressed following treatment with a checkpoint inhibitor.

Author Contributions

Conception/design: Diana Bradford, Suzanne Demko, Susan Jin, Pallavi Mishra‐Kalyani, Aaliyah R. Beckles, Kirsten B. Goldberg, Steven Lemery, Ashley Ward, Patricia Keegan, Richard Pazdur

Provision of study material or patients: Diana Bradford, Suzanne Demko, Susan Jin, Pallavi Mishra‐Kalyani, Aaliyah R. Beckles, Kirsten B. Goldberg, Steven Lemery, Ashley Ward, Patricia Keegan, Richard Pazdur

Collection and/or assembly of data: Diana Bradford, Suzanne Demko, Susan Jin, Pallavi Mishra‐Kalyani, Aaliyah R. Beckles, Kirsten B. Goldberg, Steven Lemery, Ashley Ward, Patricia Keegan, Richard Pazdur

Data analysis and interpretation: Diana Bradford, Suzanne Demko, Susan Jin, Pallavi Mishra‐Kalyani, Aaliyah R. Beckles, Kirsten B. Goldberg, Steven Lemery, Ashley Ward, Patricia Keegan, Richard Pazdur

Manuscript writing: Diana Bradford, Suzanne Demko, Susan Jin, Pallavi Mishra‐Kalyani, Aaliyah R. Beckles, Kirsten B. Goldberg, Steven Lemery, Ashley Ward, Patricia Keegan, Richard Pazdur

Final approval of manuscript: Diana Bradford, Suzanne Demko, Susan Jin, Pallavi Mishra‐Kalyani, Aaliyah R. Beckles, Kirsten B. Goldberg, Steven Lemery, Ashley Ward, Patricia Keegan, Richard Pazdur

Disclosures

The authors indicated no financial relationships.