CME
Physicians: Maximum of 1.00 AMA PRA Category 1 Credit™
Released: April 28, 2023
Expiration: April 27, 2024
KEYNOTE-057 Cohort B: Pembrolizumab for High-risk Papillary NMBIC
Elizabeth R. Plimack, MD, MS:
The current standard of care for high-risk non–muscle-invasive bladder cancer (NMIBC) is transurethral resection of bladder tumor (TURBT) followed by intravesical Bacillus Calmette-Guérin (BCG). Unresponsiveness to BCG or relapse ≤12 months indicates poor prognosis, and the standard of care in this setting is surgical resection of the bladder.23 Surgery is curative but patients are understandably averse to it, and not all patients are eligible, so novel approaches to treatment are needed.
The phase II KEYNOTE‑057 trial examined pembrolizumab in patients with BCG-unresponsive high-risk NMIBC ineligible for radical cystectomy. Cohort A of the trial, which included both carcinoma in situ (CIS) and non-CIS papillary tumors, previously reported a 3-month CR rate of 41% and median DoR of 16.2 months, with 23.1% maintaining CR for ≥24 months.24,25 However at longer term follow-up fewer than 10% of patients could be confirmed to still be in follow-up and without bladder cancer recurrence. Defining the absolute percent of patients in this cohort “cured” of their cancer at 5 years will be critical to determining if this is offered to patients with CIS as an alternative to cystectomy going forward. At ASCO GU 2023, Dr Andrea Necchi and colleagues presented an updated analysis of cohort B, a separate cohort, which includes only patients with non-CIS papillary tumors.26
Cohort B enrolled 132 patients with high-risk NMIBC unresponsive to BCG who were ineligible for or declined radical cystectomy and had received TURBT ≤12 weeks prior to trial treatment.26 Patients with papillary tumors only (high-grade Ta or any T1) without CIS were eligible. Pembrolizumab was given 200 mg IV every 3 weeks. Patients continued to receive pembrolizumab for a maximum of 35 cycles of treatment, until disease progression or recurrence.
The primary endpoints were 12‑month DFS for high‑risk NMIBC and safety. I am not sure that 12-month DFS is the most important primary endpoint. I would argue that when we treat patients with a disease curable with surgery, unless they are truly ineligible for surgery, we should see a certain cure fraction. The secondary endpoints were 12-month DFS rate for any disease (low-grade Ta, high-risk NMIBC, and PD); PFS to worsening of grade, stage, or death; PFS to muscle invasion, metastasis, or death; and OS.
KEYNOTE-057 Cohort B: Baseline Characteristics
Elizabeth R. Plimack, MD, MS:
Patients all had T1 or high-grade Ta urothelial histology per the enrollment criteria. The median age was 72 and patients had received a median of 10 prior instillations of BCG.26
KEYNOTE-057 Cohort B: Disease-Free Survival for High-Risk NMIBC
Elizabeth R. Plimack, MD, MS:
The median DFS for high-risk NMIBC was 7.7 months and the 12-month DFS rate was 43.5%.26 The estimated median DFS at 24 months and 36 months were both 34.9%.
One confounder of this analysis is the extraordinary degree of censoring beyond the primary endpoint of 12 months. Patients had to continue on the study, on the drug, receiving surveillance cystoscopies to know if they were disease free or not. So, although the estimated median DFS at 36 months is 34.9%, only 20 patients were confirmed to be disease free at that point in time. For the censored patients we just do not know. The difference between the Kaplan Meyer estimate of 34.9% and the 20 out of 132 patients, roughly 15%, is something I highlight when I talk about this with patients. Patients want to know if they are going to be one of those 20 patients confirmed disease free at 36 months.
KEYNOTE-057 Cohort B: Progression-Free Survival
Elizabeth R. Plimack, MD, MS:
There is a steady downward slope in terms of worsening of NMIBC grade or stage, or death over time with median PFS of 44.5 months. The median PFS for time to invasive or metastatic disease, or death, is 46.2 months.26 Note the shape of these curves are different than that for DFS: a steady downward slope without a plateau.
KEYNOTE-057 Cohort B: Overall Survival
Elizabeth R. Plimack, MD, MS:
The OS data caught my attention at the meeting even though a median has not been reached.26 At 12 months 96.2% of patients were still alive, but 2 important things subsequently happen. First, there is a lot of censoring, meaning patients were either lost to follow-up or have not yet reached that follow-up timepoint because of when they entered the trial. Second, there are more deaths than I would expect from this disease that is typically nonfatal unless it progresses. There are some questions as to the cause of death that were not part of the presentation, and it will be important as we follow these data further to understand this curve better.
KEYNOTE-057 Cohort B: TRAEs
Elizabeth R. Plimack, MD, MS:
It is not surprising that systemic therapy causes systemic AEs.26 Pruritus, hypothyroidism, and fatigue were the most common TRAEs but are all manageable. The incidence of grade 3/4 AEs was 14.4% and there were no treatment-related deaths.
We quote a serious AE rate of 10% to 15% to patients when they are getting pembrolizumab and consider something “serious” if it requires steroids to remediate. We carefully discuss these AEs with our patients when we go through the risks and benefits of different therapies.
KEYNOTE-057 Cohort B: Immune-Mediated AEs and Infusion Reactions
Elizabeth R. Plimack, MD, MS:
The incidence of any-grade immune‑mediated AEs and infusion reactions was 30.3%, and grade 3/4 AEs 7.6%.26 Serious AEs affected 6.8% and there were no deaths. Quality of life was improved or stable for most patients at week 45.
KEYNOTE-057 Cohort B: Clinical Implications
Elizabeth R. Plimack, MD, MS:
Based on the previous data from Cohort A, which included patients with CIS, I think a systemic approach in this setting is falling out of favor. There is not enough confirmed long‑term durability in either cohort of KEYNOTE-057. At the same time, I think we are also generating a lot of enthusiasm for novel intravesical approaches with drug‑eluting devices, intravesical chemotherapy, and intravesical viral vector therapy. At future meetings we will see data on those that will hopefully provide more incremental benefit and durable responses in these patients who, again, can be cured with surgery most of the time.
Dr Petrylak, what are your thoughts?
Daniel P. Petrylak, MD:
I think this is a population of patients, if they are truly not eligible to receive cystectomy, they are at risk for other causes of death. I would love to see a cause‑specific OS curve. I think that would give more information about what is happening in this study. If patients are dying from myocardial infarctions, they drop off this curve. If they have bad cardiovascular disease, that is a reason not to do a cystectomy.
Elizabeth R. Plimack, MD, MS:
I agree. I would think we would have seen that in the CIS cohort too because the cohorts should be the same with respect to cystectomy eligibility. It would be an interesting thing to compare to try to understand the death rate.
Neoadjuvant Pembrolizumab + Chemotherapy in MIUC: Study Design
Elizabeth R. Plimack, MD, MS:
Neoadjuvant cisplatin is the current standard of care for muscle invasive urothelial carcinoma (MIUC) requiring radical cystectomy, but many patients are cisplatin ineligible.27,28 This is a really interesting phase Ib/II study called HCRN GU14-188 examining the combination of pembrolizumab and chemotherapy in the neoadjuvant setting for patients with MIUC.29 This trial is nonrandomized, parallel-assignment, open-label, and enrolled 83 adults with cT2-4aN0M0 urothelial carcinoma eligible for surgery. Patients that were cisplatin-eligible were enrolled in cohort A and cisplatin-ineligible patients in cohort B. Cohort A received cisplatin 70 mg/m2 or 35 mg/m2 IV on Days 1 and 8 of four 21-day cycles, gemcitabine 1000 mg/m2 IV on Days 1 and 8, and pembrolizumab 200 mg IV in five 21-day cycles. Of note, cohort B received pembrolizumab plus gemcitabine (1000 mg/m2 IV on Days 1, 8, and 15 of three 28-day cycles) instead of more standard gemcitabine/carboplatin. I think that was innovative and an important thing to examine, so am glad they did that. Patients went on to receive definitive surgery with radical cystectomy or nephroureterectomy.
The current analysis, presented by Dr Jason Brown and colleagues at ASCO GU 2023, reports the primary endpoint of pathologic muscle-invasive response rate (PaIR, ≤ypT1N0) at definitive surgery, and secondary endpoints of complete pathologic (ypT0N0) response rate, definitive surgery rate, 18-month relapse-free survival (RFS), and 36-month OS.
Neoadjuvant Pembrolizumab + Chemotherapy in MIUC: Baseline Characteristics
Elizabeth R. Plimack, MD, MS:
These cohorts are not equivalent groups prognostically, and you can see that reflected in the older median age of the cisplatin-ineligible group (73 vs 64 years) and higher frequency of stage >T2 disease at diagnosis (61.5% vs 42.9%).29 Otherwise the baseline characteristics were well balanced.
Neoadjuvant Pembrolizumab + Chemotherapy in MIUC: Efficacy
Elizabeth R. Plimack, MD, MS:
The PaIR rate, which is again defined as ≤ypT1N0, was 61% among cisplatin‑eligible patients and 52% among cisplatin‑ineligible patients.29 Approximately 88% of patients in each group went on to receive surgery, and the ypT0N0 rate was similar for both groups. We tend to see ypT0N0 rates approximately 38% with neoadjuvant chemotherapy or ICI therapy in cisplatin-eligible disease.30,31 In the cisplatin-ineligible group that drops to 31% as we saw with ABACUS, which evaluated neoadjuvant atezolizumab in cisplatin‑ineligible or chemotherapy refusing cT2-4aN0M0 disease.32 These data are something to consider for cisplatin-ineligible patients eligible for cystectomy, but for whom surgery alone is probably not enough.
Neoadjuvant Pembrolizumab + Chemotherapy in MIUC: Safety
Elizabeth R. Plimack, MD, MS:
There were no new safety signals as neoadjuvant chemotherapy/immunotherapy combinations have been extensively studied in metastatic disease.29
Neoadjuvant Pembrolizumab + Chemotherapy in MIUC: Clinical Implications
Elizabeth R. Plimack, MD, MS:
I think one thing we are beginning to explore in the field is dose‑dense methotrexate, vinblastine sulfate, doxorubicin hydrochloride, and cisplatin (DD-MVAC). There are thoughts that DD-MVAC might be better than gemcitabine/cisplatin in terms of immune suppression as we examine it in the muscle‑invasive space and in randomized comparison in the VESPER trial.30
Many trials are evaluating gemcitabine/cisplatin with ICIs in the neoadjuvant setting. What was novel about this trial is that 1 arm was gemcitabine in combination with immunotherapy without cisplatin which achieved some eyebrow‑raising early results.29 I think this study is mostly important for the cisplatin‑ineligible group and I would love to see a larger, randomized trial in this population.
Dr Petrylak, what are your thoughts?
Daniel P. Petrylak, MD:
I think this study was very nicely done. The real question is how much overlap we have with neoadjuvant chemotherapy for cisplatin‑ineligible patients. Giving chemotherapy at the same time as checkpoint therapy concerns me. ICI therapy added to chemotherapy does not improve survival in the metastatic setting, so we need randomized trials to tell us what is going on in the neoadjuvant population. But the intriguing part is the cisplatin‑ineligible population, because we know single‑agent gemcitabine has an immunogenic effect. Is that synergizing with pembrolizumab in this situation? I think further studies and markers are needed to answer these questions.
JAVELIN Bladder 100 Long-term Follow-up: Avelumab Maintenance After Platinum Chemotherapy
Elizabeth R. Plimack, MD, MS:
The phase III JAVELIN Bladder 100 trial compared avelumab maintenance plus best supportive care (BSC) vs BSC alone in patients with advanced UC who did not progress on first-line chemotherapy (gemcitabine plus cisplatin or carboplatin).33,34 A significant PFS and OS benefit with avelumab led to FDA approval as maintenance therapy for patients who did not progress on first-line platinum-based chemotherapy.35 The current study, presented by Dr Srikala Sridhar and colleagues at ASCO GU 2023, reports post hoc analysis of long-term outcomes in subgroups based on first-line chemotherapy regimen and OS from the start of chemotherapy.36
JAVELIN Bladder 100 enrolled 700 patients with unresectable locally advanced or metastatic UC that did not progress on 4-6 cycles of first-line platinum-based chemotherapy. Patients were randomized to receive avelumab 10 mg/kg IV every 2 weeks plus BSC vs BSC alone. Stratification was by best response to first-line chemotherapy and metastatic site at the start of first-line chemotherapy. Treatment continued until disease progression, unacceptable toxicity, or withdrawal.
The primary endpoint is OS in all randomized patients and the PD-L1+ population and secondary endpoints are PFS (by RECIST 1.1) and safety. The post hoc analysis subgroups were defined by first-line chemotherapy regimen (cisplatin or carboplatin).
Since this patient population is highly select, we cannot compare these data to patients who start at first line.
JAVELIN Bladder 100 Long-term Follow-up: Baseline Characteristics by 1L Chemotherapy Regimen
Elizabeth R. Plimack, MD, MS:
The median age was higher among patients who previously received carboplatin, at approximately 72 years vs 66 years for patients who previously received cisplatin.36 Patients who previously received carboplatin also had a higher rate of ECOG PS of ≥1 which was expected because the selection of platinum chemotherapy is based on the PS. The best response to first-line chemotherapy was CR or PR in approximately 70% of patients across all arms and subgroups, with the remainder having SD. Recall that patients with PD to first-line chemotherapy were not eligible for JAVELIN 100 and thus not part of this analysis.
JAVELIN Bladder 100 Long-term Follow-up: OS From Start of Maintenance by 1L Chemotherapy Regimen
Elizabeth R. Plimack, MD, MS:
These data are at a median follow-up from randomization of 38.0 months in the avelumab arm and 39.6 months in the BSC alone arm, so around the 40‑month mark is where more censoring is likely to occur.36 As previously reported, there was an OS benefit with avelumab over BSC alone for patients who had previously received cisplatin or carboplatin (HR: 0.79; 95% CI: 0.611-1.020 and HR: 0.69; 95% CI: 0.516-0.925, respectively).
I think one interesting thing here is how well the control arms did. This is probably our new normal for platinum-responding patients. While the avelumab arm clearly did better in terms of median OS, it is interesting to have a more contemporary control that can be useful for other trials and trial design.
JAVELIN Bladder 100 Long-term Follow-up: PFS from Start of Maintenance by 1L Chemotherapy Regimen
Elizabeth R. Plimack, MD, MS:
For PFS we will again look around the 40‑month mark to see the cleaner curves.36 We need to be aware of the prevalence of censoring to best understand these curves, but we are seeing a clear PFS benefit in favor of maintenance avelumab compared to BSC alone in both groups of patients. The HR for patients who received first-line cisplatin, was 0.56 (95% CI: 0.446-0.713) and for patients who received first-line carboplatin, the HR was 0.48 (95% CI: 0.362-0.640).
These results are not surprising because patients receiving BSC almost always progress when they have metastatic disease, as in this case. However, I think the clinical benefit of maintenance avelumab in this setting is meaningful here.
JAVELIN Bladder 100 Long-term Follow-up: OS from Start of 1L Chemotherapy
Elizabeth R. Plimack, MD, MS:
This curve shows OS from the start of first‑line chemotherapy and I think it is a hazardous way of presenting the data.36 Note the curves are flat at the top. By definition, to enroll on this study patients could not have progressed on chemotherapy, so I would suggest that these data provide prognostic information.
The data show that patients who did not progress on first-line chemotherapy and received maintenance avelumab had a median OS of 29.7 months compared with 20.5 months with BSC alone (HR: 0.77; 95% CI: 0.636-0.921). That really is significant because for years the benchmark OS was 15 months for patients with untreated metastatic disease. The BSC alone arm is also instructive because the benchmark is now 20.5 months. Again, since these patients did not progress on first-line chemotherapy, this is a selected population; the benefits of avelumab (salvage) are probably lower in patients who progressed, but these data still help us prognosticate.
JAVELIN Bladder 100 Long-term Follow-up: OS from Start of 1L Chemotherapy by Chemotherapy Regimen
Elizabeth R. Plimack, MD, MS:
These are the OS data from start of chemotherapy separated by first-line chemotherapy regimen. Again, we see continued benefit with avelumab maintenance BSC alone regardless of first-line therapy.36
JAVELIN Bladder 100 Long-term Follow-up: Safety by 1L Chemotherapy Regimen
Elizabeth R. Plimack, MD, MS:
Since we are looking at an active treatment compared to no treatment, of course more AEs are seen with avelumab.36 The long‑term follow-up showed no new or emergent safety issues.
JAVELIN Bladder 100 Long-term Follow-up: Clinical Implications
Elizabeth R. Plimack, MD, MS:
Long-term follow-up (median follow-up ≥38 months) demonstrated PFS and OS benefits of avelumab maintenance for patients with advanced UC and no evidence of disease progression after receiving a first-line platinum-based chemotherapy regimen regardless of whether first-line chemotherapy was carboplatin- or cisplatin-based.36
I think this study reinforces the maintenance avelumab approach as a standard of care and helps us better prognosticate for our patients. In my experience, patients with a nice response to platinum‑based chemotherapy who tolerate maintenance immunotherapy have excellent quality of life, usually controlled disease, and minimal impact from treatment. How this approach fits into the rapidly changing landscape will be determined by future data.
Daniel P. Petrylak, MD:
These are intriguing data. It is great to see we have made so much progress in this disease, when, as you said, the median survival was 15 to 18 months previously. This offers hope to patients who do not progress on chemotherapy. However, we do not want to fall into a trap thinking that all patients will live 30 months. I think that is one thing that can be misinterpreted from this because most of these patients responded to first-line chemotherapy before receiving maintenance therapy. Patients with rapidly progressive disease that may not respond to first-line chemotherapy are different biologically, as well as in outcome.
Elizabeth R. Plimack, MD, MS:
I agree. We also did not see enfortumab vedotin/pembrolizumab data presented at ASCO GU, but it is an active combination potentially moving into frontline therapy for our patients. Whether better long‑term benefits are achieved starting with that combination or platinum‑based chemotherapy followed by maintenance therapy is unclear.
Daniel P. Petrylak, MD:
If we think about this in terms of what is coming in the future, we also need to look at toxicity. If the median OS is very similar with frontline enfortumab vedotin/pembrolizumab vs chemotherapy followed by maintenance avelumab, it will come down to toxicity. It also may be different in patients with visceral disease, where I would think that one would have a better chance of disease control with enfortumab-based therapy compared to chemotherapy. The incidence of long‑term neuropathy with enfortumab vedotin/pembrolizumab will need to be examined very carefully.
In terms of the EV‑302 trial (NCT04223856), do patients with nodal disease tend to respond to chemotherapy better than patients with visceral disease? Who will want to use the JAVELIN 100 type of approach rather than enfortumab vedotin/pembrolizumab? Both would use maintenance therapy but very different treatments initially to get the disease under control.
TROPHY-U-01 Cohort 2: Sacituzumab Govitecan After Checkpoint Inhibition in Platinum-ineligible Metastatic UC
Daniel P. Petrylak, MD:
Sacituzumab govitecan is a Trop-2–directed antibody-drug conjugate that received accelerated approval in patients with locally advanced or metastatic UC who progressed on prior platinum-based chemotherapy and either a PD-1 or PD-L1 inhibitor based on the results of TROPHY-U-01 cohort 1.37,38 Updated results showed an ORR of 28% and median OS of 10.9 months with a manageable safety profile. The current report, which I presented with my colleagues at ASCO GU 2023, is the primary analysis of cohort 2 which enrolled patients ineligible for platinum.39
TROPHY-U-01 cohort 2 enrolled 38 patients with mUC who progressed after ICI therapy and who were ineligible for platinum-based chemotherapy at the start of the study. Patients also had to have a creatine clearance ≥30 mL/min. Patients received sacituzumab govitecan at 10 mg/kg on Days 1 and 8 of 21‑day cycles. Treatment continued until disease progression or unacceptable toxicity.
The primary endpoint was ORR by central review (per RECIST 1.1) and secondary endpoints included duration of response (DoR), PFS by central review, clinical benefit rate, and ORR both by the central and the investigator‑run assessment.
TROPHY-U-01 Cohort 2: Baseline Characteristics and Prior Therapies
Daniel P. Petrylak, MD:
The median age was 73 years; 25 patients (66%) had visceral disease and 11 (29%) had liver metastases at baseline.39 Eight patients (21%) had 2 Bellmunt risk factors and the median number of prior anticancer regimens was 2 (range: 1-5). Prior (neo)adjuvant platinum therapy was received by 19 patients (50%), 7 patients (18%) had received prior enfortumab vedotin, and 1 (3%) previously received erdafitinib. A best response to previous therapy of CR or PR was achieved by 7 patients (19%), which is maybe a little lower overall than expected.
TROPHY-U-01 Cohort 2: Efficacy Outcomes
Daniel P. Petrylak, MD:
At a median follow-up of 9.3 months, the ORR was 32%, with PR as best response; 34% of patients had SD.39 The clinical benefit rate was 42% and median time to response was short, similar to enfortumab, at 1.4 months. The median DoR was 5.6 months among the 12 evaluable patients. The median PFS was 5.6 months, and median OS was 13.5 months. Among 13 patients who did not previously receive platinum or enfortumab, the ORR was 53.8%. The ORR was largely similar across all prespecified groups.
TROPHY-U-01 Cohort 2: Safety Outcomes
Daniel P. Petrylak, MD:
Nothing new was observed with safety outcomes in this cohort.39 Diarrhea, alopecia, and neutropenia were the predominant AEs and no treatment-related deaths were reported.
TROPHY-U-01 Cohort 2: Clinical Implications
Daniel P. Petrylak, MD:
Sacituzumab govitecan showed antitumor activity in patients with mUC who were platinum-ineligible and who progressed after ICI therapy, and had a manageable safety profile.39 The ORR was 32% in the ITT population and 53.8% in patients naive to chemotherapy and enfortumab vedotin. The safety profile was manageable and as expected with no treatment-related deaths. The ongoing phase III TROPiCS-04 trial (NCT04527991) is a randomized study evaluating sacituzumab govitecan vs chemotherapy in patients with mUC after platinum and ICI therapies.
Dr Plimack, what are your thoughts?
Elizabeth R. Plimack, MD, MS:
Sacituzumab govitecan is an active agent in urothelial cancer and is nice to have. It has a different side effect profile than enfortumab vedotin. Our practice tends to use platinum‑based chemotherapy, ICI therapy, and enfortumab vedotin before we consider this. I think this is something we would use for patients who still have good performance status after those treatments and require additional therapy.
As for the 13 patients who had not received prior platinum or enfortumab vedotin and had a 53.8% ORR, that is a rare situation to not have received those standard agents prior. Even with the interesting response rate, this strategy is not what I would go to first, especially with enfortumab vedotin plus pembrolizumab possibly moving into the first-line setting.
I think part of the issue with TROPHY-U-01 is that it is catching up to what our standard is in terms of the population treated and so it is hard for me to know where to place it. Dr Petrylak, where in your practice do you typically give sacituzumab govitecan?
Daniel P. Petrylak, MD:
I usually give it after enfortumab vedotin. But if enfortumab vedotin/pembrolizumab becomes the first‑line standard of care, what should we do in second line? Do you give sacituzumab govitecan, gemcitabine/carboplatin, or erdafitinib to FGFR‑positive patients? What is the best treatment? We need a randomized trial to tell us that. I think that is going to be very important to do.
Editor note: Enfortumab/pembrolizumab was granted accelerated approval by the FDA on 4/3/2023 for patients with locally advanced or mUC who are ineligible for cisplatin-containing chemotherapy.40
Back
