Potential Roles for PD‐1 Inhibition and Cabozantinib in Patients with Metastatic Non‐Clear Cell Renal Cell Carcinoma

Short abstract

This commentary discusses studies in metastatic non‐clear cell renal cell carcinoma and provides suggestions for choosing between treatment with nivolumab and treatment with cabozantinib.

The introduction of multitargeted tyrosine kinase inhibitors (TKIs), immune checkpoint inhibitors, and combinations of these therapeutic classes has changed the treatment landscape and improved survival for patients with metastatic clear cell (cc) renal cell carcinoma (RCC). However, 25%–30% of patients diagnosed with RCC have non‐clear cell RCC (nccRCC), which includes papillary type 1 and 2, chromophobe, mucinous tubular and spindle cell (MTSC), Xp11.2 translocation, medullary, and unclassified RCC. When compared with metastatic ccRCC, patients with metastatic nccRCC have worse clinical outcomes and distinct biology that drives tumorigenesis and have been excluded from registration trials for novel therapies 1. Accordingly, level I evidence regarding the efficacy of these drugs in patients with metastatic nccRCC is sparse, so treatment of metastatic nccRCC has simply mirrored metastatic ccRCC despite worse outcomes.

Retrospective studies have the potential to inform treatment decisions for patients with metastatic nccRCC while we await ongoing, randomized clinical trials. In this issue of The Oncologist, Dr. Chahoud and colleagues report their experience using nivolumab alone or in combination with ipilimumab or VEGF‐targeted therapies to treat patients with metastatic nccRCC at MD Anderson Cancer Center (MDACC) 2. Because their study is the third retrospective analysis of nivolumab in metastatic nccRCC, the authors also performed a meta‐analysis of these studies 3, 4. In the MDACC experience, the objective response rate (ORR) was 20.6%, with a complete response (CR) rate of 8% in a heavily pretreated cohort in which 50% of patients received nivolumab as third‐line or later treatment. Furthermore, the ORR was higher in patients who received nivolumab plus ipilimumab or VEGF‐targeted therapy compared with nivolumab alone (44.4% [4/9] vs. 9.7% [3/31]). When analyzed by histologic subtype, nivolumab produced a higher ORR in patients with unclassified RCC (44.4%, 4/9) and ccRCC with >20% rhaboid features (28.6%, 2/7). In contrast, patients with chromophobe, translocation, MTSC, or papillary type 2 RCC had an ORR of 0%. The meta‐analysis resulted in similar findings with an ORR of 18.6% in 112 patients with metastatic nccRCC.

There are some prospective data for PD‐1 inhibition in metastatic nccRCC. The recently reported KEYNOTE‐427 study included 165 patients with multiple disease subtypes; in this cohort, pembrolizumab resulted in an ORR of 24.8% 5. Papillary patients represented the preponderance of this cohort (72%), with a smaller proportion of patients with chromophobe and unclassified disease. To date, the only published randomized clinical trial data for patients with metastatic nccRCC comes from the ESPN and ASPEN trials, which compared sunitinib versus everolimus, and a single‐arm phase 2 trial of sunitinib (Table 1) 6, 7, 8. In these trials, sunitinib had an ORR of 9%, 18%, or 5%, respectively, so the ORR of 18.6% with nivolumab in the meta‐analysis of heavily pretreated patients compares favorably. In contrast, the ORR with nivolumab from the meta‐analysis in metastatic nccRCC was lower than in metastatic ccRCC (18.6 vs. 25%) 9.

Table 1.

Outcomes of clinical trials or retrospective studies for patients with metastatic non‐clear cell renal cell carcinoma

Drug(s)	Population	Design	OS	ORR, %	PFS
Sunitinib 7	Non‐clear cella (n = 57)	Single‐arm phase II	16.8 mo	5	2.7 mo
Sunitinib vs. everolimus 5	Non‐clear cella (n = 68)	Randomized phase II	16.2 vs. 14.9 mo	N/A	6.1 vs. 4.1 mo
Sunitinib vs. everolimus 6	Non‐clear cell (n = 108)	Randomized phase II	31.5 vs. 13.2 mo	18 vs. 9	8.3 vs. 5.6 mo
Cabozantinib 10	Non‐clear cell (n = 112)	Retrospective	12.0 mo	27	6.7 mo
Cabozantinib 9	Non‐clear cell (n = 30)	Retrospective	25.4 mo	14	8.6 mo
Nivolumab ± ipilimumab	Non‐clear cell (n = 40)	Retrospective	21.7 mo	21	4.9 mo
PD‐1 axis inhibitors 3	Non‐clear cella (n = 43)	Retrospective	12.9 mo	19	4.0 mo
Nivolumab 2	Non‐clear cell (n = 41)	Retrospective	NR	20	3.5 mo
Pembrolizumab 4	Non‐clear cell (n = 165)	Single‐arm phase II	N/A	24.8	N/A
Atezolizumab + bevacizumab	Non‐clear cell (n = 36)	Single‐arm phase II	21.2 mo	25	8.4 mo

^aAlso includes clear cell with >20% sarcomatoid histology.

Abbreviations: ORR, objective response rate; OS, overall survival; N/A, not available; NR, not reported; PFS, progression‐free survival.

Cabozantinib is a multitargeted TKI that has generated interest in metastatic nccRCC because of its activity against MET and AXL. Two retrospective studies have assessed the efficacy of cabozantinib in metastatic nccRCC and reported ORRs of 27% and 14% 10, 11. Because cabozantinib and nivolumab both appear to have antitumor activity in metastatic nccRCC, we can look to the subgroup analyses by histology to individualize treatment selection. Cabozantinib produced more objective responses in patients with papillary, translocation, and chromophobe RCC, whereas nivolumab produced more responses in patients with unclassified and clear cell with >20% rhomboid features RCC.

When reviewing ORRs across clinical trials and retrospective studies in metastatic nccRCC, it is worth noting the differences in patient populations between studies. The retrospective study by Dr. Chahoud and colleagues excluded patients who had ccRCC with sarcomatoid features and renal medullary carcinoma. In contrast, the ESPN trial and multiple retrospective studies included patients with sarcomatoid features in their metastatic nccRCC cohorts (Table 1). In abstracts presented at the 2019 ASCO Annual Meeting, patients with metastatic ccRCC and sarcomatoid features were shown to have exceptional responses to pembrolizumab plus axitinib and nivolumab plus ipilimumab, 59% with 11% CR and 57% with 18% CR, respectively 12, 13. These reports demonstrate how inclusion of metastatic ccRCC with sarcomatoid features would alter ORRs in analyses of patients with metastatic nccRCC. Ultimately, we need thoughtfully designed, prospective clinical trials to evaluate and compare contemporary treatments in patients with metastatic nccRCC. To highlight a few of the ongoing trials in metastatic nccRCC, PAPMET is a phase II multiarm clinical trial comparing cabozantinib, savolitinib, and crizotinib with sunitinib for patients with metastatic papillary RCC (SWOG‐1500, NCT02761057; the savolitinib and crizotininib arms were recently closed because of inferiority), CABOSUN II is a phase II clinical trial comparing cabozantinib with sunitinib in metastatic nccRCC (NCT03541902), and a single‐arm phase II clinical trial is evaluating nivolumab plus cabozantinib in metastatic nccRCC (NCT03635892).

The study by Dr. Chahoud and colleagues has several limitations that are common in retrospective studies of rare malignancies. First, patients received a heterogeneous number of prior treatments, which influences expected ORR and OS. Second, patients who received nivolumab alone or in combination with ipilimumab or VEGF‐targeted therapies were included in the analysis, but as evidenced by the registration trials, nivolumab plus ipilimumab or VEGF‐targeted therapy have different efficacy than nivolumab alone. Third, MDACC has a unique patient population, so these findings are unlikely to be generalizable to many other academic or community practices. Finally, this analysis includes multiple subgroup analyses with small numbers, so the subgroup analyses should be interpreted with caution and viewed as hypothesis generating.

Nivolumab alone, or in combination with ipilimumab, or VEGF‐targeted therapy appears to have activity in patients with metastatic nccRCC on the basis of three retrospective studies and a meta‐analysis. Subgroup analyses suggest that nivolumab and cabozantinib have differing activity among the distinct histologies that comprise metastatic nccRCC. Prospective clinical trials are needed to answer the important clinical questions raised in these studies of patients with metastatic nccRCC.

Disclosures

Sumanta K. Pal: Genentech, Aveo, Eisai, Roche, Pfizer, Novartis, Exelixis, Ipsen, Bristol‐Myers Squibb, Astellas (C/A); Neeraj Agarwal: Astellas, AstraZeneca, Bristol‐Myers Squibb, Bayer, Clovis, Eisai, Exelixis, EMD Serono, Eli Lilly & Co, Foundation Medicine, Genentech, Janssen, Merck, Novartis, Nektar, Pfizer, Pharmacyclics (C/A), AstraZeneca, Bavarian Nordic, Bristol‐Myers Squibb, Calithera, Celldex, Eisai, Exelixis, Genetech, GlaxoSmithKline, Immunomedics, Janssen, Medivation, Merck, NewLink Genetics, Novartis, Pfizer, Prometheus, Rexahn, Sanofi, Takeda, Tracon, Bayer, Clovis, EMD Serono, Eli Lilly & Co, Janssen, Nektar (RF). Andrew W. Hahn indicated no financial relationships.

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board