Abstract
On April 3, 2023, the FDA granted accelerated approval to enfortumab vedotin-ejfv (EV) plus pembrolizumab for treatment of patients with locally advanced or metastatic urothelial carcinoma who are ineligible for cisplatin-containing chemotherapy. Substantial evidence of effectiveness was obtained from EV-103/KEYNOTE-869 (NCT03288545), a multi-cohort study. Across cohorts, a total of 121 patients received EV 1.25 mg/kg (maximum of 125 mg) intravenously on days 1 and 8 of a 21-day cycle plus pembrolizumab 200 mg intravenously on day 1 of each 21-day cycle until disease progression or unacceptable toxicity. The major efficacy outcome measures were objective response rate (ORR) and duration of response (DoR) determined by blinded independent central review using RECIST v1.1. The confirmed ORR in 121 patients was 68% (95% CI: 59, 76), including 12% with complete responses. The median DoR for the 82 responders was 22 months (range: 1+ to 46+). The safety profile of the combination comprised adverse reactions expected to occur with the corresponding monotherapies, but with overall increased frequency of adverse reactions, including skin toxicity, pneumonitis, and peripheral neuropathy. The article summarizes the data and the FDA thought process supporting accelerated approval of EV + pembrolizumab, as well as additional exploratory analyses conducted by the FDA.
Introduction
Urothelial cancer (UC) accounted for approximately 16,710 deaths in 2023 in the United States (1). Standard of care for the first-line treatment of patients with locally advanced or metastatic urothelial cancer (la/mUC) is platinum-containing combination chemotherapy (2). For patients who are ineligible for cisplatin-containing therapy, carboplatin plus gemcitabine represents the most commonly used first-line therapy. Avelumab is approved in the US as first-line maintenance treatment of patients without disease progression after initial platinum-containing chemotherapy. Pembrolizumab is also approved as an acceptable option for first-line treatment of patients ineligible for any platinum-containing therapy (2). Treatments for la/mUC are not curative, and there is an unmet need for improved therapeutic options.
Enfortumab vedotin-ejfv (EV; Padcev; Astellas and Seagen) is a nectin-4-directed antibody-drug-conjugate (ADC) with a cytotoxic microtubule inhibitor payload. EV was previously approved as a single agent by the FDA for the treatment of patients with la/mUC who are ineligible for cisplatin-containing chemotherapy and have previously received one or more prior lines of therapy and for patients who have received both platinum-containing chemotherapy and either a programmed cell death protein 1 (PD-1) or programmed death ligand 1 inhibitor (PD-L1). Pembrolizumab (Keytruda; Merck) is a monoclonal antibody that binds programmed death receptor-1 (PD-1). In addition to the first-line indication for platinum-ineligible patients with la/mUC, pembrolizumab was previously approved for patients with disease progression after prior platinum-containing chemotherapy. In this approval summary, we discuss the FDA’s benefit-risk analysis of the marketing application that led to accelerated approval of EV plus pembrolizumab for treatment of adult patients with la/mUC who are ineligible for cisplatin-containing chemotherapy (summarized in Table 1) as well as additional exploratory analyses. Of note, shortly prior to this publication, the key results of the confirmatory trial EV-302 were presented (3), and the FDA granted traditional approval to EV + pembrolizumab for treatment of adult patients with la/mUC regardless of eligibility for cisplatin (4). This traditional approval will be discussed in a separate publication.
Table 1.
FDA benefit-risk assessment.
| Dimension | Evidence and Uncertainties | Conclusions and Reasons |
|---|---|---|
| Analysis of Condition | Patients with locally advanced or metastatic urothelial cancer (la/mUC) who are not eligible for cisplatin-containing therapy have limited treatment options and a poor prognosis. | Locally advanced or metastatic urothelial cancer is a life-threatening and incurable condition with unmet medical need. |
| Current Treatment Options | For cisplatin-ineligible patients with la/mUC, gemcitabine plus carboplatin is frequently used as a first-line therapy. Avelumab is approved as a maintenance therapy for patients who received first-line platinum-doublet chemotherapy and did not experience disease progression. Pembrolizumab is approved for the first-line treatment of patients with la/mUC who are not eligible for any platinum-containing chemotherapy. | There are a limited number of available treatment options for first-line treatment of cisplatin-ineligible patients with la/mUC. |
| Benefit | Patients treated with enfortumab vedotin-ejfv (EV) and pembrolizumab in pooled cohorts of study EV-103 (N = 121) showed overall response rate (ORR) of 67.8% (95% CI: 58.7, 76.0), including 12% with complete responses. The median duration of response (DoR) for the 82 responders was 22.1 months (range: 1.0 to 46.3). | The combination of EV plus pembrolizumab demonstrated an ORR and DoR that appeared superior to those of gemcitabine plus carboplatin (historic data), which represents the best available therapy for cisplatin- ineligible patients. |
| Risk and Risk Management | The combination of EV plus pembrolizumab demonstrated a higher frequency of adverse reactions compared to each agent as monotherapy. The frequency of skin toxicity, pneumonitis, and peripheral neuropathy, which are adverse reactions of special interest for EV, were more frequent in the combination versus the Cohort K monotherapy arm and historic monotherapy data. The Applicant will be required to confirm the clinical benefit of EV plus pembrolizumab in urothelial cancer. An additional post-marketing requirement was issued to further characterize peripheral neuropathy associated with EV. |
The safety profile of EV plus pembrolizumab is acceptable for the indicated population. A randomized trial, EV-302, will provide further data on the efficacy and safety of EV plus pembrolizumab in la/mUC. Additional information regarding peripheral neuropathy will be obtained from the overall EV development program. |
Clinical Pharmacology
In this application, the proposed dose of EV was 1.25 mg/kg (up to a maximum of 125 mg for patients ≥ 100 kg) administered on days 1 and 8 of a 21-day cycle to remain consistent with the pembrolizumab 21-day dosing schedule. This was different from the approved EV monotherapy dose, i.e. 1.25 mg/kg administered on days 1, 8, and 15 of a 28-day cycle (5). The proposed EV dose in the combination regimen was deemed acceptable based on the predicted similarity of PK exposures and target saturation; the similar relative dose intensity (RDI) for the 21-day regimen (75%) vs. the historical RDI (80%) at the 28-day regimen; and acceptable safety and favorable efficacy results described in this application. The combination of EV plus pembrolizumab did not substantially affect the pharmacokinetics or immunogenicity rates of either drug.
Clinical Trial Design
To support the accelerated approval, the FDA reviewed datasets from multiple cohorts for both efficacy and safety. These are summarized below and illustrated in schematic form in Figure S1 (efficacy) and Figure S2 (safety). Substantial evidence of effectiveness for the proposed indication derives from EV-103/KEYNOTE-869 (NCT03288545, hereafter referred to as EV-103), an open-label, multi-cohort study in patients with la/mUC who were ineligible for cisplatin chemotherapy and who received no prior systemic therapy for la/mUC. The approved population includes patients from the dose-escalation cohort, Cohort A, and Cohort K. The dose escalation cohort and Cohort A were single-arm cohorts where all patients received EV plus pembrolizumab, while Cohort K randomized patients 1:1 to the combination of EV plus pembrolizumab versus EV monotherapy at the same dose and schedule as in the combination (ie 1.25 mg/kg on days 1 and 8 of a 21-day cycle). In total, 121 patients (n=5, n=40, and n=76 patients in the dose-escalation cohort, Cohort A and Cohort K, respectively) were treated with EV 1.25 mg/kg (up to a maximum dose of 125 mg) on Days 1 and 8 of each 21-day cycle plus pembrolizumab 200 mg intravenously on Day 1 of each 21-day cycle. Patients were treated until disease progression or unacceptable toxicity, with no limit on the number of cycles of EV that could be received.
To isolate the single-agent activity of EV, the Applicant submitted data from Cohort K (EV monotherapy arm) of EV-103; in this arm, 73 patients were treated with EV monotherapy. To isolate the single-agent activity of pembrolizumab, the Applicant submitted external data from KEYNOTE-052 (NCT02335424), a single-arm phase 2 trial of single-agent pembrolizumab that enrolled 370 patients with la/mUC who were ineligible for cisplatin-containing therapy. The individual contributions of EV and pembrolizumab to the combination treatment were demonstrated by cross-comparison of EV-103 Cohort K (EV plus pembrolizumab) with KEYNOTE-052, and comparison between the two treatment arms in Cohort K, respectively.
The Applicant also submitted pooled safety data from 753 patients enrolled on studies of EV monotherapy.
To further aid in providing supportive evidence of efficacy in this first-line indication, supportive efficacy data was obtained from preliminary descriptive results of study EV-302 (NCT04223856), an ongoing confirmatory Phase 3 study comparing treatment of EV plus pembrolizumab to cis/carboplatin + gemcitabine chemotherapy in patients with untreated la/mUC. At the time of the FDA review, this study remained blinded to the sponsors and therefore will be described only generally here.
Statistics
The sample size for the randomized Cohort K was not based on power calculation for formal hypothesis testing, and formal statistical comparisons between the two arms of Cohort K were not planned. The primary endpoint in Cohort K was confirmed ORR, defined as the proportion of subjects with confirmed CR or PR according to RECIST v1.1. Because of the relatively small sample size in Cohort K, data on confirmed ORR were pooled from all patients treated with EV 1.25 mg/kg on Days 1 and 8 of each 3- week cycle plus pembrolizumab on Day 1 of each 3-week cycle in the dose escalation cohort, Cohort A and Cohort K. The ORR and the 2-sided 95% exact confidence intervals were estimated using Clopper-Pearson method with duration of response analyzed descriptively using Kaplan-Meier methodology.
Results
Table 2 summarizes the patient demographics and disease characteristics. The most common reasons for cisplatin ineligibility were renal insufficiency, Grade ≥ 2 hearing loss, and ECOG performance status 2. Patients with Grade ≥ 2 peripheral neuropathy were excluded from enrollment. Baseline characteristics were similar between the three pooled cohorts and were also similar between patients receiving the combination treatment versus EV monotherapy in Cohort K.
Table 2:
Demographic and baseline characteristics.
| Demographic Parameters | EV plus pembro (Combined dose escalation cohort, Cohort A, Cohort K) (N = 121) | Cohort K EV monotherapy (N=73) |
|---|---|---|
| Sex, n (%) | ||
| Male | 90 (74) | 56 (77) |
| Female | 31 (26) | 17 (23) |
| Age | ||
| Median (yrs) (Range) | 71 (51, 91) | 74 (56, 89) |
| Age Group, n (%) | ||
| < 65 years | 29 (24) | 11 (15) |
| 65-74 years | 52 (43) | 28 (38) |
| ≥ 75 years | 40 (33) | 34 (47) |
| Race, n (%) | ||
| Asian | 5 (4.1) | 6 (8) |
| Black or African American | 6 (5) | 5 (7) |
| White | 103 (85) | 55 (75) |
| Other | 2 (1.7) | 0 |
| Not reportable or unknown | 5 (4.1) | 7 (10) |
| Ethnicity, n (%) | ||
| Hispanic or Latino | 12 (10) | 2 (2.7) |
| Not Hispanic or Latino | 106 (88) | 64 (88) |
| Not reportable or unknown | 3 (2.5) | 7 (10) |
| Geographic Region, n (%) | ||
| North America | 118 (98) | 66 (90) |
| Europe | 3 (2.5) | 7 (10) |
| ECOG performance status, n (%) | ||
| 0 | 48 (40) | 28 (38) |
| 1 | 55 (45) | 35 (48) |
| 2 | 18 (15) | 10 (14) |
| Renal function based on baseline creatinine clearance a , n (%) | ||
| Normal: ≥90 mL/min | 18 (15) | 4 (5) |
| Mild decrease: ≥60 and <90 mL/min | 26 (22) | 13 (18) |
| Moderate decrease: ≥30 and <60 mL/min | 73 (60) | 53 (73) |
| Severe decrease: ≥15 and <30 mL/min | 4 (3.3) | 3 (4.1) |
| Baseline Hemoglobin A1c, n (%) | ||
| < 5.7 % | 57 (47) | 25 (34) |
| ≥ 5.7 and < 6.5 % | 54 (45) | 34 (47) |
| ≥ 6.5% | 10 (8) | 14 (19) |
| Number of Bajorin risk factors b , n (%) | ||
| 0 | 41 (34) | 26 (36) |
| 1 | 80 (66) | 47 (64) |
| Reasons for cisplatin ineligibility, n (%) | ||
| Creatinine clearancec < 60 mL/min | 73 (60) | 44 (60) |
| Grade ≥ 2 hearing loss | 16 (13) | 11 (15) |
| ECOG PS 2c | 12 (10) | 9 (12) |
| Creatinine clearancec < 60 mL/min and Grade ≥ 2 hearing loss | 12 (10) | 7 (10) |
| Creatinine clearancec < 60 mL/min and ECOG PS 2c | 6 (5) | 1 (1.4) |
| Grade ≥ 2 hearing loss and ECOG PS 2c | 1 (0.8) | 0 |
| Other | 1d (0.8) | 1e (1.4) |
| PD-L1 expression, n (%) | ||
| n | 108 | 66 |
| CPS ≥ 10 | 46 (43) | 28 (42) |
| CPS < 10 | 62 (57) | 38 (58) |
| H-score of Nectin-4 | ||
| n | 112 | 66 |
| Mean (STD) | 246 (77) | 259 (55) |
| Median | 286 | 284 |
| Q1, Q3 | 220, 300 | 240, 298 |
| Min, Max | 0, 300 | 90, 300 |
| Primary tumor location, n (%) | ||
| Upper tract | 45 (37) | 21 (29) |
| Lower tract | 76 (63) | 51 (70) |
| Both | 0 | 1 (1.4) |
| Histology type, n (%) | ||
| Urothelial cancer (UC) only | 47 (39) | 19 (26) |
| UC with squamous differentiation | 16 (13) | 7 (10) |
| UC with other histologic variants | 58 (48) | 47 (64) |
| Extent of disease at baseline, n (%) | ||
| Locally advanced | 3 (2.5) | 1 (1.4) |
| Metastatic | 118 (98) | 72 (99) |
| Time from diagnosis of locally advanced or metastatic disease to enrollmentf (months) | ||
| n | 121 | 73 |
| Mean (STD) | 2.5 (4.0) | 3.6 (10) |
| Median | 1.5 | 1.6 |
| Min, max | 0.3, 35 | 0.5, 86 |
| Disease stage at baseline, n (%) | ||
| IIIA | 2 (1.7) | 1 (1.4) |
| IV | 45 (37) | 23 (32) |
| IVA | 19 (16) | 11 (15) |
| IVB | 55 (45) | 38 (52) |
| Metastatic disease sitesg at baseline, n (%) | ||
| Bone | 27 (22) | 21 (29) |
| Lymph Nodes | 85 (70) | 56 (77) |
| Lung | 56 (46) | 30 (41) |
| Liver | 27 (22) | 13 (18) |
| Metastasis category, n (%) | ||
| Lymph node only disease | 17 (14) | 12 (16) |
| Visceral metastases | 102 (84) | 60 (82) |
| Not applicableh | 2 (1.7) | 1 (1.4) |
Source: U.S. Food and Drug Administration. Clinical Review – BLA 761137, Supplement 018 (enfortumab vedotin-ejfv) and BLA 125514, Supplement 136 (pembrolizumab).
Creatinine clearance was estimated using the Cockroft-Gault formula based on the last non-missing serum creatinine measurement before the first dose of study treatment.
Bajorin risk factors comprise visceral metastases (bone, lung, liver) and Karnovsky performance status < 80 % (> ECOG 2)
Cisplatin eligibility was determined based on creatinine clearance and ECOG performance status at screening, which were not identical to baseline values shown above.
One patient in Cohort A was considered cisplatin ineligible by the investigator due to a solitary kidney.
One patient in the Cohort K EV monotherapy arm was considered cisplatin ineligible by the investigator due to age and Grade 1 hearing loss.
Calculated from the date of locally advanced or metastatic disease for subjects whose disease is locally advanced or metastatic at the time of enrollment.
A subject may have had metastatic disease in more than one location.
Subjects had locally advanced disease without metastasis to lymph nodes or distant organs.
Efficacy:
The confirmed ORR in the 121-patient efficacy population from EV-103 was 67.8% (95% CI: 58.7, 76.0), including 12% with complete responses, as determined by blinded independent central review (BICR) (Table 3). The median DoR for the 82 responders was 22.1 months (range: 1.0+ to 46.3+) with a median follow-up of 18.5 months (95% CI 15.4, 21.3). Response rates were similar across exploratory subgroups (age, sex, PD-L1 status, primary disease site of origin, etc). The confirmed ORR in the EV-103 Cohort K EV monotherapy arm was 45.2% (95% CI: 33.5, 57.3), including 4.1% with complete responses; and DoR was 13.2 months (range: 1.5+ to 16.4+). The confirmed ORR of pembrolizumab monotherapy from KEYNOTE-052 was 28.9% (95% CI 24.3-33.8); and DoR was 33.4 months (range: 1.4+ to 60.7+). Preliminary ORR results from the confirmatory trial (EV-302) were consistent with the results from patients receiving combination treatment in EV-103 (not shown). As noted earlier, ORR results from EV-302 have since been published (3,4).
Table 3:
Efficacy results per blinded independent central review assessment.
| EV plus pembro (Combined dose escalation cohort, Cohort A, Cohort K) (N = 121) | Cohort K EV monotherapy (N=73) | |
|---|---|---|
| Primary efficacy endpoint | ||
| Confirmed overall response ratea | ||
| n (%) | 82 (67.8) | 33 (45.2) |
| (95% CI)b | (58.7, 76.0) | (33.5, 57.3) |
| CR, n (%) | 15 (12.4) | 3 (4.1) |
| PR, n (%) | 67 (55.4) | 30 (41.1) |
| Duration of response,c months | ||
| Median (95% CId) | 22.1 (17.15, NE) | 13.2 (6.14, 15.97) |
| Range | 0.95+, 46.32+ | 1.51+, 16.39+ |
| DoR ≥ 6 months,e n (%) | 57 (69.5) | 15 (45.5) |
| DoR ≥ 12 months,e n (%) | 30 (36.6) | 4 (12.1) |
Source: U.S. Food and Drug Administration. Clinical Review – BLA 761137, Supplement 018 (enfortumab vedotin-ejfv) and BLA 125514, Supplement 136 (pembrolizumab).
Abbreviations: CI: confidence interval; CR: complete response; PR: partial response; DoR: duration of response.
Response assessed per blinded independent central review according to Response Evaluation Criteria in Solid Tumor v1.1.
Computed using the Clopper-Pearson method (Clopper 1934).
As estimated using the Kaplan-Meier method.
Calculated using the complementary log-log transformation method (Collett, 1994).
As observed on study.
Safety:
The primary safety population for EV-103 was the same as the 121-patient efficacy population. The 73 patients receiving EV monotherapy in Cohort K, who were planned to receive the same dose and schedule of EV as patients in the combination arm, are included here for comparison. The duration of exposure to overall study treatment in EV-103 was longer for patients receiving the combination than for those receiving EV monotherapy (median 9 vs 6 months). The median cumulative dose of EV received was also greater in patients receiving the combination. However, the median dosing intensity of EV was lower in patients receiving combination treatment, presumably because delays and EV dose reductions were more common with combination treatment (Table 4).
Table 4:
Summary of exposure and safety data in EV-103.
| EV plus pembro (Combined dose escalation cohort, Cohort A, Cohort K) (N = 121) | Cohort K EV monotherapy (N=73) | |
|---|---|---|
| Exposure summary | ||
| Number of cycles with any drug, median (min, max) | 10 (1, 36) | 8 (1, 33) |
| Duration of treatment with any drug, months; median (min, max) | 9 (0.6, 33) | 6 (0.5, 27) |
| EV dose intensity,a mg/kg/cycle; median (min, max) | 1.9 (0.9, 2.6) | 2.3 (1.2, 2.7) |
| Cumulative EV dose, mg; median (min, max) | 1500 (94, 6180) | 1045 (180, 3872) |
| Overall safety summary | ||
| Patients with any AR, n (%) | 121 (100) | 73 (100) |
| Patients with any grade 3-5 AR, n (%) | 103 (85) | 57 (78) |
| Patients with any serious AR, n (%) | 61 (50) | 31 (42) |
| Patients with fatal ARb, n (%) | 7 (6) | 5 (7) |
| Patients with treatment discontinuationc due to AR, n (%) | 27 (28) | 18 (28) |
| Patients with treatment delay due to AR, n (%) | 93 (77) | 33 (45) |
| Patients with EV dose reductions, n (%) | 59 (49) | 40 (41) |
| Selected adverse reactions | ||
| Rash,d n (%) | ||
| All grades | 86 (71) | 37 (51) |
| Max grades 3-4 | 24 (20) | 6 (8) |
| Fatal | 1 (0.8) | 0 |
| Peripheral neuropathy,e n (%) | ||
| All grades | 79 (65) | 43 (59) |
| Max grade 2 | 55 (45) | 22 (30) |
| Max grade 3 | 4 (3.3) | 3 (4.1) |
| Pneumonitis, n (%) | ||
| All grades | 11 (9) | 3 (4.1) |
| Max grades 3-4 | 4 (3.3) | 0 |
| Fatal | 1 (0.8) | 0 |
| Myositis, all grades, n (%) | 4 (3.3) | 0 |
| Myasthenia gravis, n (%) | ||
| All grades | 3 (2.5) | 0 |
| Fatal | 1 (0.8) | 0 |
Source: U.S. Food and Drug Administration. Clinical Review – BLA 761137, Supplement 018 (enfortumab vedotin-ejfv) and BLA 125514, Supplement 136 (pembrolizumab).
Abbreviations: AR, adverse reaction
EV dose intensity is defined as the actual dose per unit of time that a subject received over the entire treatment period (time from first dose to day 21 of last treatment cycle, regardless if death occurs before the end of cycle). For reference, the intended dose intensity is defined in the protocol as 2.50 mg/kg/3 weeks cycle (ie 1.25 mg/kg on days 1 and 8).
Six deaths (5%) were considered to be fatal adverse reactions. These included sepsis (1.6%), bullous dermatitis (0.8%), myasthenia gravis (0.8%), and pneumonitis/ILD (0.8%).
All study treatment stopped
Includes: blister, conjunctivitis, dermatitis, dermatitis bullous, dermatitis exfoliative generalized, erythema, erythema multiforme, exfoliative rash, palmar-plantar erythrodysesthesia syndrome, pemphigoid, rash, rash erythematous, rash macular, rash maculo-papular, rash papular, rash pruritic, rash vesicular, skin exfoliation, stomatitis
Includes: dysesthesia, gait disturbance, hypoesthesia, muscular weakness, paresthesia, peripheral motor neuropathy, peripheral sensorimotor neuropathy, peripheral sensory neuropathy
The combination of EV and pembrolizumab was associated with similar categories of adverse reactions as have been observed with the respective monotherapies (5,6). However, patients treated with combination therapy had a higher frequency of clinically significant adverse reactions (ARs, Table 4) and laboratory abnormalities (Table S1) than patients treated with EV monotherapy in Cohort K. Of ARs of special interest for EV, as defined by the Warnings and Precautions in the existing product label (5), the frequency of skin toxicity, pneumonitis/interstitial lung disease (ILD), and peripheral neuropathy were increased in the combination versus the Cohort K monotherapy arm and historic monotherapy data. Myositis and myasthenia gravis also occurred more frequently in the combination arm, although the total number of events was low.
FDA evaluated safety in the ECOG performance status (PS) 2 population (n=18/121 or 14.8% of patients receiving combination treatment), given that the PS 2 population might be more representative of a treated population in a real-world (i.e. non-trial) setting involving patients deemed cisplatin ineligible. Although the total number of patients with ECOG PS 2 was relatively small in EV- 103, there was no apparent difference in treatment exposure or overall safety between this subgroup and the rest of the studied population (Table S2).
FDA conducted additional analyses of peripheral neuropathy in both the 121 patients receiving combination therapy in EV-103, and in 753 patients treated with EV monotherapy pooled across multiple trials. Incidence of new peripheral neuropathy and Grade ≥ 2 peripheral neuropathy in patients who did not previously have neuropathy was approximately constant over time in both the EV-103 safety population (Table S3) and the EV monotherapy pool (Table S4), with incidence of 20-40% in each 3-month time period. Resolution of neuropathy was similar between the EV-103 safety population and the EV monotherapy pool: approximately 86% of patients who had developed peripheral neuropathy on these studies had residual peripheral neuropathy at last evaluation of neuropathy, including 40% with residual Grade ≥ 2 peripheral neuropathy (Table S5). Strategies of treatment discontinuation in response to peripheral neuropathy in patients receiving EV + pembrolizumab on EV-103 were heterogeneous. Seven patients discontinued both treatments at the same time because of peripheral neuropathy, while 16 patients discontinued EV alone due to peripheral neuropathy. These latter 16 patients continued pembrolizumab for a median duration of 8 months (range: 1 to 18+ months) after discontinuing EV.
Regulatory Insights
This approval represents the first approval for a combined ADC and immune checkpoint inhibitor and represents a new first-line treatment option for cisplatin-ineligible patients with la/mUC. Study EV-103 included an EV monotherapy arm to demonstrate contribution of components, however EV monotherapy is not currently approved for first-line treatment of la/mUC.
The review team determined that the combination of EV plus pembrolizumab demonstrated efficacy that was better than available first-line therapy for cisplatin-ineligible patients based on substantially higher ORR than both carboplatin plus gemcitabine (generally 35-45%) (7-9) and pembrolizumab (28.9% with 95% CI of 24.3-33.8 on KEYNOTE-052) (10); and a duration of response that compared favorably to carboplatin plus gemcitabine (generally 4-8 months) (8,9). Avelumab, which was approved during the conduct of EV-103 for maintenance treatment of patients with at least stable disease on platinum-based chemotherapy, may be associated with an additional ORR of up to 9.7% in addition to the ORR of platinum-based chemotherapy alone (11), although direct extrapolation of the JAVELIN Bladder 100 study results to EV-103 is not possible given differences in eligibility and other study characteristics. The ORR of patients receiving EV plus pembrolizumab in EV-103 was also greater than that observed with either EV monotherapy (EV monotherapy arm of Cohort K in EV-103) or pembrolizumab monotherapy (KEYNOTE-052), supporting the contributions of both EV and pembrolizumab to the combination. The median DoR of patients receiving EV plus pembrolizumab was supportive when compared with that of EV monotherapy.
Acknowledging the limitations inherent in non-randomized comparisons and challenges in accurately assessing DoR, the median DoR of 22.1 months in patients treated with the combination in EV-103 appeared shorter than the 33.4 months (range: 1.4+ to 60.7+) observed with pembrolizumab monotherapy in KEYNOTE-052 (10). There may be numerous potential explanations for this, including differences in various study-specific factors across trials. Additionally, some initial responders to EV + pembrolizumab may have responded to EV alone without durable immunologic response. For some patients, the combined toxicity of the combination may have caused early stopping of treatment due to intolerance or toxicity.
Given that safety was qualitatively similar to the known safety profiles of the respective monotherapies, no new categories of safety signals were identified for inclusion as Warnings & Precautions or boxed warnings in labeling. Overall, the review team considered the increased incidence and severity of toxicity of combination treatment to be acceptable in the context of the demonstrated efficacy. The dosing strategy used in EV-103 was to administer both treatments until disease progression or until intolerable toxicity requiring discontinuation of one or both treatments. Current recommendations for dose reductions and discontinuations are provided in the product labels for the respective treatments. Whether any alternative dosing strategy might mitigate toxicity while preserving efficacy would need to be explored in a dedicated randomized trial.
FDA conducted additional analyses of peripheral neuropathy given that this was the most frequent adverse reaction leading to treatment delay, dose reduction, and treatment discontinuation of EV (5); and because even persistent Grade 2 neuropathy can substantially impact patients’ functioning and quality of life. An exploratory analysis by FDA did not identify a threshold of cumulative exposure to EV after which the incidence of peripheral neuropathy increases, unlike what has been observed with platinum agents (12,13). The patterns of incidence and resolution of neuropathy appeared similar between patients treated with EV plus pembrolizumab and EV monotherapy, and some patients were able to receive pembrolizumab for a prolonged duration after EV was discontinued due to peripheral neuropathy. These observations support that peripheral neuropathy in patients with the combination may be mediated primarily by EV. One reason that peripheral neuropathy may have been more frequent with combination treatment vs EV monotherapy is that improved efficacy of the combination led to continuation of the treatment, leading to increased cumulative exposure to EV.
Peripheral neuropathy may become increasingly impactful as the EV plus pembrolizumab regimen is studied in larger patient populations and in less refractory disease settings. As such, the FDA and the Applicant agreed on a post-marketing requirement to conduct an integrated safety analysis of ongoing trials across the EV development program to further characterize the known serious risk of peripheral neuropathy in relation to prolonged exposure to EV in the settings of perioperative treatment of localized UC and in first-line treatment of la/mUC.
Conclusion
The FDA’s evaluation of the submitted efficacy and safety data supports that EV plus pembrolizumab has a clinically meaningful treatment effect, with durable responses and an acceptable safety profile in patients who have a significant unmet medical need (Table 1). The magnitude of the ORR with supportive DoR was substantially improved over that of available therapies for this population and thus met the requirement for accelerated approval. Submission of the final results for the confirmatory study (EV-302), which randomizes patients with la/mUC to either platinum-based chemotherapy or EV plus pembrolizumab and includes both cisplatin eligible and ineligible patients, is a post-marketing requirement. Of note, study EV-302 was fully enrolled at approximately the same time as the application for accelerated approval was submitted, which is consistent with current FDA recommendations regarding timely completion of confirmatory trials (14). A second post-marketing requirement was issued to further characterize peripheral neuropathy in patients receiving this combination.
Supplementary Material
Footnotes
Disclosure of Potential Conflicts of Interest: SH completed work on this publication while he was an employee at the US Food and Drug Administration. At the time of publishing, he is an employee at Astellas. EB completed work on this publication while he was an employee at the US Food and Drug Administration. At the time of publishing, he is an employee at Merck. The remaining authors report no financial interests or relationships with the commercial sponsors of any products discussed in this report. CCB completed work on this publication while she was an employee at the US Food and Drug Administration. At the time of publishing, she is an employee at Vertex Pharmaceuticals.
This is a U.S. Government work. There are no restrictions on its use.
References
- 1.SEER Cancer Stat Facts: Bladder Cancer. Bethesda, MD: National Cancer Institute; 2023. [Google Scholar]
- 2.NCCN: NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®), Version 3; 2023. <https://www.nccn.org/professionals/physician_gls/pdf/bladder.pdf>. [Google Scholar]
- 3.Powles TB, Perez Valderrama B, Gupta S, Bedke J, Kikuchi E, Hoffman-Censits J, et al. LBA6 EV-302/KEYNOTE-A39: Open-label, randomized phase III study of enfortumab vedotin in combination with pembrolizumab (EV+P) vs chemotherapy (Chemo) in previously untreated locally advanced metastatic urothelial carcinoma (la/mUC). Annals of Oncology 2023;34:S1340 doi 10.1016/j.annonc.2023.10.106. [DOI] [Google Scholar]
- 4.2023 FDA approves enfortumab vedotin-ejfv with pembrolizumab for locally advanced or metastatic urothelial cancer, December 15, 2023. <https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-enfortumab-vedotin-ejfv-pembrolizumab-locally-advanced-or-metastatic-urothelial-cancer>. [Google Scholar]
- 5.FDA: US Prescribing Information: enfortumab vedotin-ejfv. 2023. [Google Scholar]
- 6.FDA: US Prescribing Information: pembrolizumab. 2023. [Google Scholar]
- 7.De Santis M, Bellmunt J, Mead G, Kerst JM, Leahy M, Maroto P, et al. Randomized phase II/III trial assessing gemcitabine/carboplatin and methotrexate/carboplatin/vinblastine in patients with advanced urothelial cancer who are unfit for cisplatin-based chemotherapy: EORTC study 30986. J Clin Oncol 2012;30(2):191–9 doi 10.1200/jco.2011.37.3571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Holmsten K, Jensen NV, Mouritsen LS, Jonsson E, Mellnert C, Agerbæk M, et al. Vinflunine/gemcitabine versus carboplatin/gemcitabine as first-line treatment in cisplatin-ineligible patients with advanced urothelial carcinoma: A randomised phase II trial (VINGEM). Eur J Cancer 2020;127:173–82 doi 10.1016/j.ejca.2019.08.033. [DOI] [PubMed] [Google Scholar]
- 9.Park I, Kim B-S, Lim HY, Kim H-J, Lee HJ, Choi YJ, et al. Gemcitabine plus carboplatin versus gemcitabine plus oxaliplatin in cisplatin-unfit patients with advanced urothelial carcinoma: a randomised phase II study (COACH, KCSG GU10–16). European Journal of Cancer 2020;127:183–90 doi 10.1016/j.ejca.2019.08.034. [DOI] [PubMed] [Google Scholar]
- 10.Balar AV, Castellano DE, Grivas P, Vaughn DJ, Powles T, Vuky J, et al. Efficacy and safety of pembrolizumab in metastatic urothelial carcinoma: results from KEYNOTE-045 and KEYNOTE-052 after up to 5 years of follow-up. Ann Oncol 2023;34(3):289–99 doi 10.1016/j.annonc.2022.11.012. [DOI] [PubMed] [Google Scholar]
- 11.Powles T, Park SH, Voog E, Caserta C, Valderrama BP, Gurney H, et al. Avelumab Maintenance Therapy for Advanced or Metastatic Urothelial Carcinoma. New England Journal of Medicine 2020;383(13):1218–30 doi 10.1056/NEJMoa2002788. [DOI] [PubMed] [Google Scholar]
- 12.Saif MW, Reardon J. Management of oxaliplatin-induced peripheral neuropathy. Ther Clin Risk Manag 2005;1(4):249–58. [PMC free article] [PubMed] [Google Scholar]
- 13.Staff NP, Cavaletti G, Islam B, Lustberg M, Psimaras D, Tamburin S. Platinum-induced peripheral neurotoxicity: From pathogenesis to treatment. J Peripher Nerv Syst 2019;24 Suppl 2(Suppl 2):S26–s39 doi 10.1111/jns.12335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fashoyin-Aje LA, Mehta GU, Beaver JA, Pazdur R. The On- and Off-Ramps of Oncology Accelerated Approval. New England Journal of Medicine 2022;387(16):1439–42 doi 10.1056/NEJMp2208954. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.