Welcome to the CCO Site

Thank you for your interest in CCO content. As a guest, please complete the following information fields. These data help ensure our continued delivery of impactful education. 

Become a member (or login)? Member benefits include accreditation certificates, downloadable slides, and decision support tools.

Submit

Key Studies in Genitourinary Cancers: CCO Independent Conference Highlights of the 2021 ASCO Annual Meeting
  • CME

Daniel P. Petrylak, MD
Elizabeth R. Plimack, MD, MS
Released: September 7, 2021

Key Studies in RCC

KEYNOTE-564: Study Design

Elizabeth R. Plimack, MD, MS:
At ASCO 2021, we saw the presentation of the first interim analysis of the KEYNOTE-564 trial as part of the plenary session, which showcases the most high-impact data at ASCO.1-2 This trial was a randomized, double-blind phase III study of adjuvant pembrolizumab vs placebo in patients with resected ccRCC who had a nephrectomy ≤12 weeks before randomization and had no previous systemic therapy (N = 994). Enrolled patients had an Eastern Cooperative Oncology Group (ECOG) PS of 0 or 1 and were required to have a tissue sample for PD-L1 assessment. Eligible patients were randomized 1:1 to receive either pembrolizumab at 200 mg every 3 weeks or placebo every 3 weeks for approximately 1 year.

The primary endpoint of the KEYNOTE-564 study was disease-free survival (DFS), and a key secondary endpoint was OS, which is probably more clinically relevant for most of us. The investigators also looked at safety. Although the adverse event (AE) profile with pembrolizumab has been well described across many studies, toxicity should be weighed differently in the adjuvant setting for healthy patients undergoing adjuvant treatment for a disease that may never metastasize even without treatment.

KEYNOTE-564: Baseline Characteristics

Elizabeth R. Plimack, MD, MS:
The trial arms were well balanced. Approximately 70% of patients were male, and 85% of patients had an ECOG PS of 0. In total, 84% of patients had no sarcomatoid features, and most (74%-77%) had a PD-L1 CPS ≥1.

Patients in the KEYNOTE-564 study were enrolled and stratified based on 3 disease risk categories: Patients with pT2, grade 4 or with sarcomatoid features, or any-grade pT3 and no lymph node involvement and metastasis free (M0) were considered intermediate/high risk; patients with pT4 or any pT and lymph node positive and M0 were considered high risk; and patients with no evidence of disease (NED) after complete resection of oligometastatic sites ≤1 year from nephrectomy were considered M1 NED. In both arms of the trial, most patients enrolled (86%) had M0 intermediate/high-risk disease. Patients with M0 high-risk disease made up 7% to 8% of the patients enrolled, and patients in the M1 NED category accounted for 5.8% of patients enrolled.

KEYNOTE-564: DFS

Elizabeth R. Plimack, MD, MS:
The KEYNOTE-564 trial met the primary endpoint of an improvement in DFS. The median DFS was not reached in either arm of the trial at a median follow-up of 24.1 months. The 12-month DFS rate was 86% with adjuvant pembrolizumab vs 77% with placebo, and the 24-month DFS was 77% with pembrolizumab vs 68% with placebo. The HR for DFS was 0.68 (95% CI: 0.53-0.87; P = .0010).

Within the cohort of patients identified for this study, more than one half of patients in the placebo arm at 2 years have not had disease recurrence. Most patients who do not recur after 2 years will never recur, and will eventually be considered to be cured. This group of patients is similar to any group overtreated with adjuvant therapy—a high risk at baseline resulted in adjuvant therapy, but they already were cured of their disease.

The conundrum is that we do not know which patients with high-risk disease are cured vs which will have disease recurrence. Biomarkers to determine that will be critical to increase the therapeutic index of any adjuvant therapy, including in this setting.

KEYNOTE-564 DFS by Subgroup

Elizabeth R. Plimack, MD, MS:
Subgroup analyses are often criticized because studies are not designed specifically to answer this question. However, we can leverage subgroup analyses to help determine whether to recommend a specific therapy to a specific patient in clinical practice with characteristics that are outlined in these analyses.

The benefit of adjuvant pembrolizumab in KEYNOTE-564 was consistent across the subgroups examined, including age, sex, ECOG PS status, geographic region, and metastatic staging. Most patients (75%) had a PD-L1 CPS ≥1, and these patients received more of a benefit from adjuvant pembrolizumab treatment than those with a lower CPS score. The HR for those with a CPS ≥1 was 0.67 (95% CI: 0.51-0.88) vs 0.83 (95% CI: 0.45-1.51) for those with a CPS <1.

The data for patients with CPS ≥1 are compelling and, when considering adjuvant pembrolizumab, suggest that a high PD-L1 CPS portends a greater DFS benefit compared with placebo.

KEYNOTE-564: Interim OS

Elizabeth R. Plimack, MD, MS:
The median OS was not reached for either treatment arm at a median follow-up of 24.1 months. The 12-month OS rate for both arms was similar, at approximately 98%, and the 24-month OS rate was 96.6% with adjuvant pembrolizumab vs 93.5% with placebo. The HR for OS was 0.54 (95% CI: 0.30-0.96; P = .0164).

KEYNOTE-564: Safety

Elizabeth R. Plimack, MD, MS:
When looking at safety in the KEYNOTE-564 trial, it is important to remember that some proportion of these patients may not have recurrent RCC and, therefore, may not need adjuvant treatment. This presentation of the data showed the incidence of AEs, which is helpful, but what was not described yet very critical in this space is chronicity of toxicity.

We know that adrenal AEs, thyroid AEs, joint pain, and rashes all can be chronic and may not be reversible with steroids or other applicable treatment. I would like to see longer-term follow-up with information on chronic AEs and whether they resolve over time.

For all-cause AEs, 20% of patients in the pembrolizumab group and 2% of patients in the placebo group experienced an AE that led to treatment discontinuation.

In total, 18.9% of patients in the pembrolizumab arm experienced a grade 3-5 treatment-related AE vs 1.2% of patients in the placebo group.

In the context of metastatic disease, this AE profile would be considered favorable compared with other available treatment options. But, again, in a setting where patients are otherwise healthy, this AE profile is important to keep in mind.

KEYNOTE-564: Immune-Mediated AEs

Elizabeth R. Plimack, MD, MS:
The most common immune-mediated AEs for patients receiving pembrolizumab were hypothyroidism (21%) and hyperthyroidism (12%), and the incidence of these was low in the placebo group (3.6% and 0.2%, respectively). Grade 3 or 4 immune-mediated AEs were uncommon (<2% incidence for any grade), with severe skin reaction and type 1 diabetes being most frequent. A small percentage of patients in the pembrolizumab group (7.4%) and the placebo group (0.6%) used high-dose corticosteroid treatment for their immune-mediated AEs.

KEYNOTE-564: Conclusions

Elizabeth R. Plimack, MD, MS:
In the KEYNOTE-564 study, most patients enrolled had intermediate/high-risk disease. The study met the primary endpoint for DFS (HR: 0.68), and the survival benefit was consistent across the subgroups examined. Overall, patients with higher PD-L1 CPS (≥1) tended to do better than those with a lower CPS score (<1).

The OS data are not mature and remind us that almost all study participants are surviving after treatment for kidney cancer, which is great. Even though I worry about patients who are resected but at high risk for recurrence, I would not offer adjuvant pembrolizumab to my patients with high-risk disease until we see that we are significantly improving OS outcomes for this patient population. Although these data are exciting, my reasoning for this is that some toxicities with immune checkpoint inhibitor therapy can be chronic, and we know from these data that more than one half of patients did not recur at 2 years, even in the placebo group. Some of those patients will never recur, so the burden of overtreatment is present.

In the metastatic setting, combination therapies with an immune checkpoint inhibitor and a TKI are very effective, so it is theoretically possible that, for those patients destined to recur, we can achieve the same OS if we treat them when they metastasize as we would if we treated them early. If adjuvant pembrolizumab only delays recurrence and does not cure patients, I think it is unlikely that we will see an OS benefit in the long run, because the first-line therapies we have for metastatic disease are so effective. I am looking forward to seeing long-term follow-up with OS data so we can better understand if there is an overall benefit for adjuvant pembrolizumab vs waiting to treat patients if they develop metastatic disease.

Daniel P. Petrylak, MD:
I agree that we should consider the potential risk of chronic toxicities and the 20% discontinuation rate along with the positive DFS data. For example, in this study, there was a 2% incidence of pneumonitis with pembrolizumab therapy, and pneumonitis can become a chronic problem. In my practice, we have seen interstitial pneumonitis occur 1 year after a patient started treatment with an immune checkpoint inhibitor, and it can be a chronic toxicity that can have a significant impact on a patient’s quality of life. So, I completely agree with your analysis.

I also am looking forward to seeing the long-term OS data so we can understand more about how adjuvant pembrolizumab can benefit our patients. With the balance of efficacy and safety in mind, I think it would be worth pursuing a trial of adjuvant pembrolizumab in a high-risk group of patients rather than the mostly intermediate/high-risk group that has been described in this trial.

KEYNOTE-426 Final Analysis: Study Design

Elizabeth R. Plimack, MD, MS:
At ASCO 2021, 4-year follow-up data from the KEYNOTE-426 study were presented.3-4 KEYNOTE-426 was a randomized, open-label phase III study for patients with newly diagnosed or metastatic ccRCC who had no prior systemic therapy for advanced disease. Patients (N = 861) were randomized to receive pembrolizumab plus axitinib or sunitinib monotherapy. Pembrolizumab treatment was given for 2 years, and axitinib treatment could be continued beyond 2 years in patients for whom it was appropriate. Sunitinib treatment was given until disease progression.

This study previously met its dual primary endpoints of OS and PFS, and the data presented at ASCO 2021 provided longer-term follow-up for efficacy in this population.3,5-6 We are seeing unprecedented long-term duration of efficacy in patients with metastatic RCC that was not seen before the use of novel combination therapies, so it is important to continue to follow the data over time and to understand the long-term implications for OS and how it affects patient outcomes.

KEYNOTE-426 Final Analysis: Baseline Characteristics

Elizabeth R. Plimack, MD, MS:
The patient baseline characteristics were well balanced between treatment arms.3 The median age was approximately 61 years, and almost three quarters of the enrolled patients were male. PD-L1 CPS of ≥1 was reported in approximately 60% of patients. In terms of the type of disease, 83% of patients previously received a nephrectomy, and most (73%-77%) had 2 or more sites of metastasis. It also is important to note that approximately 30% of patients had International Metastatic RCC Database Consortium (IMDC) favorable-risk disease.

KEYNOTE-426 Final Analysis: Subsequent Therapy

Elizabeth R. Plimack, MD, MS:
This updated analysis for KEYNOTE-426 included data on the type of subsequent therapy patients received after discontinuing study treatment, if any. Most patients in this trial—349 in the pembrolizumab plus axitinib arm (81.4%) and 385 in the sunitinib arm (90.6%)—discontinued treatment. In general, patients discontinue cancer treatment because either their disease progresses or they die—and although that may be true in some cases, with immune checkpoint inhibitor therapy combined with a VEGF TKI, there also is a group of patients who discontinue treatment with 1 or both agents because they are doing well and have achieved disease control that can be maintained off therapy.

Many healthcare professionals have a goal of achieving treatment-free survival for our patients with kidney cancer, where we hope to be able to treat a patient for a specific period of time and that they will continue to do well after we have stopped treatment. I think all of us who treat kidney cancer in the clinic have a collection of patients who are in this category.

Keeping that in mind, of the 349 patients who received pembrolizumab plus axitinib and who discontinued therapy, 58% received subsequent therapy. For the 385 patients who received sunitinib and then discontinued therapy, 73% received subsequent treatment. The fact that we have fewer patients who discontinued treatment with pembrolizumab plus axitinib who went on to receive subsequent therapy suggests that some of these patients did not need additional therapy and have achieved long-term disease control.

When you look at the type of subsequent treatment received, it is what would be expected. It is more difficult to determine the best next-line therapy for patients who already have received both a VEGF TKI and IO. After progression on sunitinib, the next regimen should be an IO-based regimen, and—in fact—that is what happened. In our clinical practice, after progression on pembrolizumab plus axitinib, we recommend either a clinical trial or cabozantinib as a strong VEGF inhibitor, and that is reflected in the study data.

KEYNOTE-426 Final Analysis: PFS and OS in the ITT Population

Elizabeth R. Plimack, MD, MS:
OS and PFS were the primary endpoints of KEYNOTE-426, and both were met at the first interim analysis with an HR for PFS of 0.69 (95% CI: 0.57-0.84) and an HR for OS of 0.53 (95% CI: 0.38-0.74).5 The HR has changed somewhat over time because we are losing fewer patients toward the end of the survival curves, because once patients do well, they tend to do well for a longer period of time.3 After a median follow-up of 42.8 months, the HR for OS was 0.73 (95% CI: 0.60-0.88; P <.001), and the HR for PFS was 0.68 (95% CI: 0.58-0.80; P <.0001). Both HRs were statistically significant in favor of pembrolizumab plus axitinib vs sunitinib.

Looking at the landmark OS analysis is important because it allows us to compare across clinical trials. Normally, cross-trial comparisons are not recommended, but in this era, there are multiple clinical trials of various combination treatments with nearly identical design, and comparing landmark OS allows us to look at similar data points across these clinical trials. This type of comparison will help us to determine which combination regimen to recommend for our patients with advanced RCC.

For KEYNOTE-426, the 36-month OS rate is 63% for pembrolizumab plus axitinib vs 54% for sunitinib. KEYNOTE-426 and CheckMate 214, which compared nivolumab plus ipilimumab with sunitinib, have the longest follow-up and have similar 36-month OS rates.7

In addition, the 24-month OS in KEYNOTE-426 is on par with what has been reported in the CheckMate 9ER and CLEAR trials with nivolumab plus axitinib and pembrolizumab plus lenvatinib vs sunitinib, respectively.8-9 So, the landmark analyses for OS all are being hit in the same way—the only difference is the length of follow-up based on when each study was started and the time course over which each was conducted.

KEYNOTE-426 Final Analysis: Response

Elizabeth R. Plimack, MD, MS:
For our patients with kidney cancer, we want them to live as long possible and live as well as possible, but when a patient first comes to our clinic, we also try to get a response to treatment as quickly as possible. We know responding patients do better overall, and many patients are symptomatic from the bulk of their tumor, which can present as night sweats or specific localized pain from the tumor. So, achieving a quick response and getting those tumors to shrink is clinically relevant to our patients and in our practice.

In the KEYNOTE-426 trial, the median duration of response (DoR) in the combination arm was 23.6 months vs 15.3 months in the sunitinib arm. Most patients achieved a partial response, and it is notable here that pembrolizumab plus axitinib has a high overall response rate (ORR) of 60.4% vs 39.6% with sunitinib. The ORR with pembrolizumab plus axitinib was higher than what was reported with ipilimumab plus nivolumab in the ITT population of the CheckMate 214 trial (39.1% vs 32.6% with sunitinib).7 In general, the IO/VEGF TKI combination therapies stand out in terms of initial overall response.

The low rate of primary progression is another aspect of treatment that favors a combination IO/VEGF TKI therapy approach. Only 11.3% of patients had progressive disease as best response with pembrolizumab plus axitinib treatment vs 17% with sunitinib. Keeping patients from progressing and keeping their disease controlled also is a valuable endpoint clinically.

KEYNOTE-426 Final Analysis: Efficacy by IMDC Risk

Elizabeth R. Plimack, MD, MS:
When discussing treatment options for patients with newly diagnosed advanced RCC, IMDC risk criteria are important to mention. These criteria were developed in the TKI era before immune checkpoint inhibitors were approved, but they still are relevant for helping us to separate patients into categories when we are thinking about first-line treatment.

In the ITT population for KEYNOTE-426, the 42-month OS rate was 57.5% with pembrolizumab plus axitinib vs 48.5% with sunitinib. However, in patients with favorable-risk disease, the 42-month OS rate was similar in both arms at 72.3% with pembrolizumab plus axitinib vs 73.0% with sunitinib. For those with intermediate/poor-risk disease, the 42-month OS in the combination arm was 50.6% vs 37.6% in the sunitinib arm. These data suggest that patients with favorable-risk disease do not achieve any additional OS benefit when we add an immune checkpoint inhibitor to a VEGF TKI. This could be for multiple reasons. One could be that using a first-line VEGF TKI followed by an immune checkpoint inhibitor provides an OS similar to the OS when using these agents in combination up front. It also could be that immune checkpoint inhibitors in this group of patients do not have much of an impact, and I think that hypothesis is supported by the low rate of response in patients with favorable-risk disease receiving ipilimumab and nivolumab,10 which is a pure IO regimen.

Either way, if you are considering an IO/VEGF TKI combination for a patient with favorable-risk disease, it should not be for an OS benefit, but more so for the hope of a more rapid response and the potential for long-term disease control.

KEYNOTE-426 Final Analysis: Most Common Treatment-Related AEs

Elizabeth R. Plimack, MD, MS:
With a combination therapy, one would expect a worse incidence of AEs, but there were similar numbers of deaths (0.9% vs 1.6%) and grade 3-5 treatment-related AEs (67.8% vs 63.8%) in the KEYNOTE-426 study for the pembrolizumab plus axitinib arm and the sunitinib arm. The toxicity does not really change our treatment strategy in this setting of advanced disease, because combination therapy is much more effective than single agents.

However, similar to my comments on AEs above, it would be nice to see data on the chronicity of the AEs associated with this IO/VEGF TKI combination approach, particularly as we are extending the time patients are receiving therapy and extending our patients’ lives.

KEYNOTE-426 Final Analysis: Conclusions

Elizabeth R. Plimack, MD, MS:
The KEYNOTE-426 trial has the longest follow-up described for patients receiving an IO/VEGF TKI combination approach. There were no new safety signals, durable responses, and a good OS in this updated analysis. We know that in this study, patients stopped pembrolizumab after 2 years on therapy, and now that we have data at the 4-year follow-up, we are seeing patients still benefiting from this combination, even though they are not still receiving pembrolizumab—and that is compelling. We can give therapy for a period of time, stop it, and still maintain benefit.

Daniel P. Petrylak, MD:
I think the data from KEYNOTE-426 are showing that the combination of pembrolizumab plus axitinib is making a huge difference, but I like the analysis that you gave of the favorable-risk patients: that it may not be necessary to give these patients IO at this point in their disease.

Elizabeth R. Plimack, MD, MS:
That is controversial—it is not something everyone agrees with—but I think it is important to point out, because it has been consistent for the various IO/VEGF TKI combination trials.

For patients with favorable-risk disease in my practice, we often recommend observation, and we consider metastasectomy with the goal of delaying the need for a systemic therapy. For a patient with favorable-risk disease who needs systemic therapy but is worried about AEs, it is reasonable to start with a VEGF TKI alone, but that is the only context in which I would give a single agent. For a patient who really wants to be aggressive for the best long-term benefit or who needs a response even if they are favorable risk, the combination of pembrolizumab plus axitinib would be my choice.

CheckMate 9ER: Study Design

Elizabeth R. Plimack, MD, MS:
The CheckMate 9ER study is another trial combining a VEGF TKI with an immune checkpoint inhibitor.11-12 This was a randomized phase III trial of nivolumab plus cabozantinib vs sunitinib at standard dosing for patients with previously untreated advanced or metastatic RCC. Patients (N = 651) were treated until disease progression (per Response Evaluation Criteria in Solid Tumors v1.1) or intolerable toxicity. Of note, this study did not mandate cessation of nivolumab at a certain point in time.

The primary endpoint in CheckMate 9ER was PFS, and the secondary endpoints were OS, ORR, and safety.

CheckMate 9ER: Baseline Characteristics

Elizabeth R. Plimack, MD, MS:
The groups in this trial were relatively evenly balanced with respect to IMDC risk, the number of organ sites with metastases, and the sum of the target lesion diameter. More patients had liver metastases in the nivolumab plus cabozantinib group compared with the sunitinib group (22.6% vs 16.5%). Overall, more patients with intermediate and poor risk were enrolled than we saw in the KEYNOTE-426 study, which had approximately one third with favorable risk.3 Here, approximately 23% of patients had favorable risk in both treatment arms.11

CheckMate 9ER: PFS and OS by IMDC Risk Subgroup

Elizabeth R. Plimack, MD, MS:
In this analysis, PFS and OS were assessed by IMDC risk. The HRs for PFS for patients with favorable-risk and intermediate-risk disease were 0.58 (95% CI: 0.36-0.93) and 0.58 (95% CI: 0.45-0.76), respectively. The HR for PFS for patients with poor-risk disease was even better at 0.36 (95% CI: 0.23-0.56). These data suggest that patients with poor-risk disease do best with the addition of an immune checkpoint inhibitor as part of their treatment and that sunitinib monotherapy would not be a good option.

Similar to what was reported in KEYNOTE-426, in the CheckMate 9ER trial, there was no observed OS benefit with combination IO/VEGF TKI therapy vs single-agent sunitinib for patients with metastatic ccRCC and favorable-risk disease. The HR for OS in this group was 0.94 (95% CI: 0.46-1.92) for nivolumab plus cabozantinib vs sunitinib. However, patients with either intermediate- or poor-risk disease do benefit from nivolumab plus cabozantinib vs sunitinib, especially those with poor-risk disease. The HR for OS in patients with poor risk was 0.45 (95% CI: 0.27-0.76) in favor of combination therapy, suggesting again that patients in this risk group should receive a combination IO/VEGF TKI or IO/IO.

CheckMate 9ER: Overall Response and Best Overall Response by IMDC Risk Subgroup

Elizabeth R. Plimack, MD, MS:
The combination of nivolumab plus cabozantinib confers a better ORR even in the favorable-risk group, and this is where we see the true benefit of combination IO/VEGF TKI therapy. In the favorable-risk group, the ORR was 66% (95% CI: 54.3%-76.8%) with nivolumab plus cabozantinib compared with 44% (95% CI: 32.7%-56.6%) with sunitinib. However, there were similar complete response rates in the favorable-risk group (approximately 10%) regardless of the treatment arm. So, even though there was no OS benefit in this population, there was a benefit when looking at response rates. If you are considering nivolumab plus cabozantinib for patients with favorable-risk disease, it should be for those who need a response to therapy.

For patients with intermediate or poor risk, the response rates begin to decrease when comparing across the risk groups for the combination approach, but the ORR improves for each group when comparing nivolumab plus cabozantinib vs sunitinib. The ORR in the intermediate-risk group was 55.9% (95% CI: 48.4%-63.1%) and 28.7% (95% CI: 22.4%-35.8%) for nivolumab plus cabozantinib and sunitinib, respectively. For the poor-risk group, the ORR was 37.7% (95% CI: 25.6%-51.0%) and 10.3% (95% CI: 4.2%-20.1%). Overall, response rates were higher with nivolumab plus cabozantinib than with single-agent sunitinib in all risk groups.

CheckMate 9ER: PFS and OS by Other Subgroups

Elizabeth R. Plimack, MD, MS:
The PFS and OS benefit of nivolumab plus cabozantinib over sunitinib was consistent across all subgroups evaluated (site of metastases, target lesion diameter, and sum of target lesion diameters at baseline).

Patients with poor-risk disease tend to have more sites of disease, more visceral disease, and higher tumor burden, and those are the patients who benefit the most from the nivolumab plus cabozantinib combination. But when we look at it granularly based on those parameters, we just see consistency of benefit with the combination over sunitinib.

CheckMate 9ER: Conclusion

Elizabeth R. Plimack, MD, MS:
Overall, CheckMate 9ER is another positive study for an IO/VEGF TKI combination approach. This study met its endpoints (PFS and OS), clearly showing that the combination of nivolumab plus cabozantinib is effective, particularly for patients with poor- or intermediate-risk disease. This study was a bit different in that the analysis separated poor risk from intermediate risk and helped us to further understand how important it is to give an IO component together with a VEGF inhibitor for patients with poor-risk disease.

Provided by Clinical Care Options, LLC

Contact Clinical Care Options

For customer support please email: customersupport@cealliance.com

Mailing Address
Clinical Care Options, LLC
12001 Sunrise Valley Drive
Suite 300
Reston, VA 20191

Supported by educational grants from
Exelixis
Merck Sharp & Dohme Corp.

Leaving the CCO site

You are now leaving the CCO site. The new destination site may have different terms of use and privacy policy.

Continue

Cookie Settings