Modest Activity With Durvalumab/Savolitinib Combo in Papillary RCC Not Enough to Warrant Further Studies

The overall response rate achieved with with the combination of savolitinib and durvalumabin in patients with metastatic papillary renal cell carcinoma in the phase II CALYPSO study did not meet the prespecified criterion for further investigation of the regimen.

The combination of savolitinib and durvalumab (Imfinzi) induced durable responses in approximately one-fourth of patients with metastatic papillary renal cell carcinoma (RCC), according to findings from the phase II CALYPSO study.1 However, the overall response rate (ORR) achieved in the trial did not meet the prespecified criterion for further investigation of the regimen.

The overall study population of 41 evaluable patients, who were either treatment-naïve or previously treated, 11 patients (27%) achieved a response to the combination of savolitinib and durvalumab, with all 11 responses being a partial response (PR), said senior study author Cristina Suárez, MD, PhD, in a presentation during the 2019 Genitourinary Cancers Symposium. The study design called for an ORR ≥50% to warrant further investigation of the combination.

The prognosis for patients with advanced papillary RCC is poor. Currently available therapies, which were developed mainly for the treatment of clear cell RCC, have limited efficacy.

Clinical guidelines recommend either enrollment into a clinical trial, which is the preferred strategy, or sunitinib (Sutent) in patients with papillary RCC. Phase II trials indicate that VEGF TKIs, mTOR inhibitors, and MET inhibitors have some clinical activity in papillary RCC.

“However, overall survival is approximately 12 months, with some exceptions in selected populations,” said Suárez, of Hospital University Vall d’Hebron—Vall d’Hebron Institute of Oncology in Barcelona, Spain. “More robust evidence for treatment options in papillary RCC is required.”

Savolitinib is a potent and selective small molecule inhibitor of c-MET kinase, said Suárez. MET is a membrane-bound receptor TKI that is activated by its ligand, hepatocyte growth factor, and triggers cell proliferation. Durvalumab is a humanized monoclonal antibody that inhibits PD-L1.

As monotherapy, savolitinib was associated with an ORR of 7% in a phase II study of 111 patients with locally advanced papillary RCC.2 A previous phase I dose-escalation trial identified durvalumab at a 1500-mg dose every 4 weeks and savolitinib at 600 mg daily as safe, with no dose-limiting toxicities.3

CALYPSO is a multi-arm study that includes patients with different RCC histologies: clear cell RCC, sarcomatoid, and papillary RCC. A total of 42 patients with measurable metastatic papillary RCC were enrolled, the data from which were presented at the 2019 Genitourinary Cancers Symposium. One patient experienced disease progression prior to study therapy. To be eligible for enrollment, patients could be either naïve to VEGF inhibitors or refractory to treatment.

Savolitinib was given as monotherapy at 600 mg/day during a 4-week lead-in, after which durvalumab at 1500 mg was added to therapy. Treatment continued until disease progression or unacceptable toxicity. The primary endpoint was ORR; secondary endpoints were progression-free survival (PFS), overall survival (OS), duration of response, best response at 24 weeks, and safety.

The median patient age was 62 years and all patients had an ECOG performance status of 0 (39%) or 1 (61%). International mRCC Database Consortium prognosis was good in 12 patients (29%), intermediate in 26 (63%), and poor in 3 (7%). Metastatic sites included bone in 17%, central nervous system in 5%, liver in 20%, and lung in 44%. Thirty-five patients (85%) had undergone prior nephrectomy. About two-thirds (68%) of patients were treatment-naïve, 22% had 1 previous antitumor treatment, and 10% had 2 prior lines of therapy.

In the overall study cohort, in addition to the 11 responders, 16 (39%) patients achieved stable disease (SD; 95% CI, 24%-55%). Twenty-two of the 41 evaluable patients (54%) had any decrease in tumor burden. Of the 11 patients with objective response (95% CI, 14%-43%), an interim analysis showed a duration of response approaching 6 months. Three patients enrolled had baseline scans only and were not evaluable for response.

In the subset of 28 patients with previously untreated disease, the ORR was 32% (95% CI, 16%-52%), with all 9 responses being PRs. Twelve (43%) of the previously untreated patients had SD, and 2 patients had only baseline scans and were not evaluable.

At a median follow-up of 6.9 months, in an interim analysis, the investigator-assessed median PFS was 5.3 months (95% CI, 1.5-12.0) and median OS was not reached (95% CI, 7.5-not reached); however, the interim OS data are immature.

Eight of the 41 patients were PD-L1 positive, defined as >25% immune component using the SP263 antibody, and 27 were PD-L1 negative. Ten of the 41 patients were MET positive, defined as ≥3+ in ≥50% tumor cells as analyzed by immunohistochemistry (IHC), and 25 were MET negative. Six patients were not assessable or did not have enough tumor tissue available for an analysis of either biomarker.

“PD-L1 and MET biomarker expression did not clearly correlate with outcome,” said Suárez. When responses were assessed by PD-L1 and MET status, the best overall response was a PR in 3 of the 8 (38%) patients who were PD-L1 positive and 2 of the 10 (20%) who were MET positive. One additional patient (13%) who was PD-L1 positive and 5 (50%) who were MET positive had stable disease.

The combination was tolerable; the most prominent all-grade adverse events were nausea (n = 16), edema (n = 12), and fatigue (n = 9), with the majority being grade 1/2, said Suárez. The most frequent grade 3 AE was edema, which occurred in 4 patients. One patient had a grade 4 toxicity (transaminitis); there were no grade 5 toxicities.

Discussant Tracy L. Rose, MD, MPH, assistant professor of medical oncology, at the University of North Carolina at Chapel Hill, said that the 4-week lead-in of savolitinib—likely the less active of the 2 agents—may have diminished the ORR, referring to the 7% ORR with savolitinib monotherapy in the previous trial. Furthermore, accurate selection of MET-driven tumors is critical to the success of savolitinib.

“High MET protein levels by IHC may not correlate with response,” she said, wondering if the outcome would have been different with a different definition of MET [positivity]. In addition, the possibility exists that MET positivity predicts lack of response to immune checkpoint inhibition. “These questions are not answered here,” said Rose.

Finally, it is unclear that combination treatment is better than sequential given no signal of synergy and toxicity appears additive, she concluded.


  1. 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; abstr 545).
  2. Choueiri TK, Plimack E, Arkenau H-T, et al. Biomarker-based phase II trial of savolitinib in patients with advanced papillary renal cell cancer. J Clin Oncol. 2017;35(26):2993-3001. doi: 10.1200/JCO.2017.72.2967.
  3. Vari S, Szabados B, de Liaño Lisa AG, et al. A phase ib study of durvalumab (MEDI4736) in combination with savolitinib in patients with metastatic renal cell carcinoma. Presented at: the 2nd International Congress on Clinical Trials in Oncology and Hemato-Oncology; February 19-20, 2018; Berlin, Germany.

View more from the 2019 Genitourinary Cancers Symposium