Pembrolizumab/Axitinib Favored as Frontline Standard in mRCC

Thomas Powles, MBBS, MRCP, MD

The combination of pembrolizumab (Keytruda) and axitinib (Inlyta) is favored as the frontline standard of care for patients with metastatic clear cell renal cell carcinoma (RCC) following data from the pivotal phase III KEYNOTE-426 trial, said Thomas Powles, MBBS, MRCP, MD.

In the long-anticipated trial, investigators randomized patients to receive either the combination of pembrolizumab and axitinib or sunitinib (Sutent) monotherapy. Results indicated a 47% reduction in the risk of death and a 31% reduction in the risk of disease progression or death. The median progression-free survival (PFS) was 15.1 versus 11.1 months with the combination versus single-agent sunitinib, respectively,¹ and the median overall survival (OS) was not yet reached in either arm.

The benefit was observed irrespective of International Metastatic Renal Cell Carcinoma Database Consortium [IMDC] risk status or PD-L1 expression. Moreover, the 12- and 18-month OS rates with the combination and sunitinib were 89.9% versus 78.3% and 82.3% versus 72.1%, respectively. Additionally, the 12- and 18-month PFS rates were 59.6% versus 41.1% and 46.2% versus 32.9% with the combination compared with sunitinib, respectively.

“In kidney cancer, we haven’t had a lot of big studies with big survival advances,” said Powles. “We certainly haven't had one with a hazard ratio as low as 0.53. As I see it, we’ll [choose the regimen] with the best survival data, which at the moment is axitinib and pembrolizumab.”

Based on these data, the FDA granted a priority review designation to the combination in February 2019 as a frontline treatment for patients with advanced RCC.

Moreover, in the phase II CALYPSO trial, patients with metastatic papillary RCC derived a modest benefit, with the combination of the MET inhibitor savolitinib and the PD-L1 inhibitor durvalumab (Imfinzi), which was shown to induce durable responses approaching 6 months in 11 of the 41 patients enrolled.²

Although the overall response rate (ORR) of 27% fell short of the trial’s prespecified criteria for further exploration, the data helped establish the importance of incorporating immune checkpoint inhibition into a treatment approach beyond patients with clear cell disease, said Powles.

OncLive: What was the rationale for the KEYNOTE-426 trial?

In an interview with OncLive, Powles, director of Barts Cancer Institute, highlighted key takeaways from these trials and discussed how they impact the ever-changing landscape of mRCC.Powles: In frontline metastatic clear cell kidney cancer, there are 2 main classes of agents: the angiogenic agents like axitinib or sunitinib, and the immune checkpoint inhibitors like pembrolizumab. Both classes have activity as monotherapy in kidney cancer. Sunitinib, axitinib, and pembrolizumab all show single-agent response rates around 35%.

What were the key findings?

In a phase Ib trial, the combination of axitinib and pembrolizumab was shown to bump those response rates up to above 60%. The provocative phase I data suggest that by combining the drugs together, you may get more activity. Therefore, we wanted to conduct a definitive randomized phase III study to explore whether or not the combination could beat a benchmark control like sunitinib in frontline metastatic clear cell kidney cancer.KEYNOTE-426 showed that the combination of pembrolizumab and axitinib outperformed sunitinib from a survival perspective with a hazard ratio of 0.53—that translates to a 47% reduction in the risk of death, which is a major finding for survival.

By combining these 2 drugs, we're getting a powerful antiangiogenic and anti-immune effect, which translates to patients living significantly longer. The original sunitinib trials only showed PFS advantage. The VEGF-targeted therapies did move the field forward over a period of years, but we didn't have this big wave of survival in the frontline setting. This is going to change the way we think about [treating patients with] kidney cancer in the future.

There are other aspects that one needs to consider within the context of this trial. It's not all about survival; the coprimary endpoint was PFS. There, we also saw a very impressive delay in progression at 15 months versus 11 months with axitinib plus pembrolizumab and sunitinib, respectively. The 11-month PFS with sunitinib is important because it shows that the control arm performed well. The response rate was significantly higher with axitinib and pembrolizumab. Moreover, the complete response (CR) rate was 6% with the combination.

Is PD-L1 an effective biomarker in mRCC?

Are there any forthcoming biomarkers?

Were there any unanticipated toxicities or challenges with combined VEGF and PD-1 inhibition?

We need more mature data, as this is an interim analysis. The rate of grade 3/4 toxicities was similar between the 2 [arms]. About 8% of patients needed to stop the combination, while about 10% of patients needed to stop sunitinib. Overall, you can see a strong rationale for the combination—an excellent survival signal, high response rates, and tolerability. The combination is practice changing.The results of the KEYNOTE-426 study showed that the drugs were effective, irrespective of IMDC risk group and PD-L1 status. We've seen data before in kidney cancer that has been quite contradictory in this environment. There's a lot of uncertainty around the biomarker. Clearly, if you're giving the VEGF-targeted therapy and the immune therapy together, you're getting the antiangiogenic effect, which may dilute the effect of the PD-L1 biomarker. As it currently stands, PD-L1 does not appear to be the right biomarker to select patients for the combination of pembrolizumab and axitinib.The biomarker story has really been led by the data from IMmotion150 and IMmotion151 with the combination of bevacizumab (Avastin) and atezolizumab (Tecentriq). Those trials used RNA sequencing and FoundationOne CDx to look for gene signatures and mutations associated with response and resistance. We can see distinct angiogenic signatures and T-effector gene signatures associated with response to VEGF-targeted therapies and immune therapies, respectively. Those gene signatures are available to us, but that work is being validated. It’s a huge contribution to the future because we may be able to select patients [for therapy according to their signature]. In the future, we may be able to personalize therapy.Toxicity is an important part of the discussion. These drugs are going to change practice, and we need to learn how to give them. The toxicity profile is acceptable, but there are key areas that we need to look at carefully. Historically, we’ve been able to stop VEGF-targeted therapy if diarrhea and transaminitis occur and then watch things settle. We need to be prudent about immunotherapy rather than VEGF-targeted therapy, because it can be more catastrophic if left to drift by itself.

Could you shed light on the results of the phase II CALYPSO trial?

We need to go through a period of education and training to get on top of the immune-related toxicity, which seems prominent when [immunotherapy is] given in combination with VEGF-targeted therapy. We've treated many patients with the combination of axitinib and pembrolizumab. My experience is that the vast majority of patients do very well, but we do need to make sure that we optimize dosing and educate patients correctly, particularly during the beginning of therapy. If patients do have problems on the drugs, they will know who to contact and we’ll know how to treat [the adverse events] by discontinuing the drug temporarily, introducing steroids, and monitoring liver function tests.In CALPYSO, we looked at the combination of the PD-L1 inhibitor durvalumab and the MET inhibitor savolitinib in patients with metastatic papillary RCC.

We're beginning to see some preliminary data showing that PD-1/PD-L1 inhibitors have some activity in papillary renal cancer. We know from previous studies with savolitinib that MET is a potentially active target in this environment. By combining these 2 drugs, we hope to get additive effects in terms of disease control—that’s an area of huge unmet need. In this study, we showed response rates of 27%, which are high compared with sunitinib and other drugs that have been looked at in this environment.

[Patients were not selected based on a biomarker], but we did look at the PD-L1—positive cohort as well as the MET-positive cohort. We showed that response rates are higher in the PD-L1–positive cohort, at more than 30%. MET didn't really seem to work as a biomarker for this group. This suggests that a lot of the activity was probably driven by the PD-L1 inhibitor durvalumab.

There are some toxicities associated with the drug. It causes some fatigue and edema. This is an area of unmet need. We can change the way we treat [patients with] papillary renal cancer by being a bit smarter than we have been in the past. Originally, we used VEGF-targeted therapies because that's what we were treating kidney cancer with. Clear cell kidney cancer is VEGF driven, but papillary kidney cancer is not. Now, we're thinking smarter and using checkpoint inhibitors.

What other compelling studies were presented at the 2019 Genitourinary Cancers Symposium?

We saw some excellent data with pembrolizumab, some nice data with bevacizumab and atezolizumab, and some OK data with durvalumab and savolitinib. I hope we’re starting a new chapter in papillary cancer in which we don't treat it exactly the same as we would clear cell kidney cancer. Hopefully, we can start refining treatment by making it more targeted and improve the outcome of this group of patients who are still lagging behind the clear cell cohort.The JAVELIN Renal 101 study looked at avelumab (Bavencio) and axitinib. That study shows similar response rates and PFS to the data [I presented on KEYNOTE-426]. It doesn't show the survival advantage that we are going to need to see [moving forward]. The 30-month follow-up on ipilimumab (Yervoy) and nivolumab (Opdivo) were also presented, which showed a consistent hazard ratio for survival with high CR rates of 10%.

How will you approach sequencing now?

The main highlight for me is the data with pembrolizumab and axitinib. We know ipilimumab and nivolumab also has a survival advantage. If atezolizumab/bevacizumab and avelumab/axitinib want to come into this space, we'll need to see those survival advantages.We look at survival as a very important marker of choice. When you have survival advantages, cross-trial comparisons can be quite difficult. Therefore, there is going to be a genuine choice and there are different determinants that go into making that choice. A 47% reduction in the risk of death is pretty significant. We have not seen that before. That figure is going to be compelling for the community. Ipilimumab and nivolumab has a hazard ratio of 0.63 in intermediate- and poor-risk patients; that is also impressive. The community will have to decide how to choose between these agents.

References

1. Powles T, Plimack ER, Stus V, et al. Pembrolizumab (pembro) plus axitinib (ax) versus sunitinib as first-line therapy for locally advanced or metastatic renal cell carcinoma (mRCC): phase III KEYNOTE-426 study. J Clin Oncol. 2019;37(suppl 7; abstr 543). doi: 10.1200/JCO.2019.37.7_suppl.543.
2. Powles T, Larkin JMG, Patel P, et al. A phase II study investigating the safety and efficacy of savolitinib and durvalumab in metastatic papillary renal cancer (CALYPSO). J Clin Oncol. 2019;37(suppl 7; abstr 545). doi: 10.1200/JCO.2019.37.7_suppl.545.