Immunotherapy rechallenge in metastatic renal cell carcinoma: a meta-analysis of randomized clinical trials

Abstract

Introduction

The rapid integration of immune checkpoint inhibitor (ICI) based combination therapies in first-line treatment of metastatic renal cell carcinoma (mRCC) is raising questions about next-line treatments and outcomes of ICI rechallenge.

Methods

We performed a meta-analysis using the results from the phase-III RCTs, CONTACT-03 and TiNivo-2, to evaluate the effect of adding a PD-1/PD-L1 inhibitor rechallenge to a vascular endothelial growth factor (VEGF) tyrosine kinase inhibitor (TKI) vs a VEGF TKI alone in patients with mRCC who had progressed on ICI based combination therapy or ICI monotherapy.

Results

A total of 865 patients with mRCC were included in this meta-analysis. The meta-analysis showed no difference in progression free survival between ICI plus VEFG TKI combination and TKI monotherapy groups (HR = 0.96, 95% CI, 0.76-1.21; P = .75) in patients previously progressing on ICI. Furthermore, adding ICI to anti-VEGF TKI therapy was not associated with improve overall survival (HR = 1.06, 95% CI, 0.89-1.25; P = .52).

Conclusion

In summary, PD-1/PD-L1 inhibitor rechallenge with a VEGF TKI in mRCC does not improve survival outcomes and should not be used in patients with prior progression to ICI. Understanding both intrinsic and acquired resistance mechanisms to ICI therapy is crucial for developing effective initial and sequential immunotherapy strategies in RCC.

Immune checkpoint inhibitor (ICI)-based combinations have become the cornerstone of treatment for metastatic renal cell carcinoma (mRCC) over the past decade. Their rapid integration into clinical practice as first-line treatments and in the adjuvant setting after nephrectomy for patients with high risk of recurrence has resulted in uncertainty in treatment sequencing for patients whose disease has progressed after treatment with ICIs, raising questions about whether rechallenge with ICIs leads to improved clinical outcomes in subsequent lines of therapy. Retrospective analyses and single-arm trials have shown encouraging activity for ICI-TKI combinations following progression on ICI; two randomized clinical trials (RCTs), CONTACT-03 and TiNivo-2, were phase 3 trials designed to answer this question.^1,2

We performed a meta-analysis using the data from the phase-III RCTs, which compare the combination of ICI and vascular endothelial growth factor (VEGF) tyrosine kinase inhibitor (TKI) against VEGF TKI monotherapy in patients with mRCC who previously progressed on ICI-based combination therapy or ICI monotherapy. This systematic review was conducted in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.³ The literature search was conducted in the PubMed database for relevant studies between January 01, 2018, and December 31, 2024. The keywords searched were “((“kidney cancer”) OR (“renal cell carcinoma”)) AND ((“immune checkpoint inhibitors”) OR (“nivolumab”) OR (“atezolizumab”) OR (“pembrolizumab”) OR (“avelumab”) OR (“durvalumab”) OR (“toripalimab”)) AND ((“rechallenge”) OR (“retreatment”) OR (“re-administration”) OR (“readministration”) OR (“restart”) OR (“progression”) OR (“progressive”))”. The PRISMA diagram of search results and the risk of bias of included studies were shown in Figures S1 and S2; see online supplementary material for a color version of this figure. The primary endpoints were progression-free survival (PFS) and overall survival (OS). The meta-analysis conducted the generic inverse-variance method under a random-effects model to assess the risks of PFS and OS, using the hazard ratio (HR) along with its 95% confidence interval (CI). Review Manager software version 5.3 was used to conduct all statistical analyses. Statistical significance thresholds were established at 0.05 for overall effect tests and 0.10 for heterogeneity tests. ­Heterogeneity among studies was quantified using the I² coefficient.

The detailed flowchart outlining the study selection process is presented in Figure S1, See online supplementary material for a color version of this figure. A total of 865 patients (434 patients in experimental arms) with mRCC were included in this meta-analysis.^1,2 All patients had to have progressed on ICI. In the CONTACT-03 trial, all patients had progressed on ICI-based therapy as their last treatment. However, the TiNivo-2 trial also included patients who may have received VEGFR TKI after progressing on ICI before enrolling. The baseline characteristics of the included trials are shown in Table 1. The meta-analysis of the studies showed that (Figure 1) there is no difference in PFS between ICI plus VEFG TKI and TKI monotherapy groups (HR = 0.96, 95% CI, 0.76-1.21; P = .75). No significant difference in PFS was observed between the treatment arms across the IMDC risk groups (HR = 1.19, 95% CI, 0.79-1.81; P = .42 in the IMDC favorable risk group, HR = 0.91, 95% CI, 0.74-1.12; P = .37 in the IMDC intermediate risk group, and HR = 1.34, 95% CI, 0.90-1.98; P = .15 in the IMDC poor risk group) and patients who received study treatment as second-line and third-line (HR = 1.08, 95% CI, 0.87-1.35; P = .47 in the second-line and HR = 1.02, 95% CI, 0.79-1.31; P = .89 in the third-line). Similarly, there was no OS benefit with the ICI plus anti-VEGF TKI combination compared to the anti-VEGF TKI monotherapy (HR = 1.06, 95% CI, 0.89-1.25; P = .52).

Table 1.

Baseline characteristics of the trials.

Trials	Experimental arm and control arm	Number of patients	Median age (range) years	Male sex (%)	Non-clear histology (%)	IMDC favorable risk (%)	IMDC intermediate risk (%)	IMDC poor risk (%)	ICI as most recent line of therapy (%)	Rechallenge as second line (%)	Rechallenge as third line (%)	Median follow-up time, (range), months	Median PFS for ITT (95% CI), (months)	PFS for ITT HR (95% CI)	Median OS for ITT (95% CI) (months)	OS for ITT HR (95% CI)	ORR % CR %
CONTACT-03	Cabozantinib plus atezolizumab	263	62 (20-85)	204 (78)	30 (11)	49 (19)	172 (65)	41 (16)	263 (100)	144 (54.7)	119 (45.2)	15.6 (10.7-19.3)	10.6 (9.8-12.3)	1.03 (0.83-1.28)	25.7 (21.5-NE)	0.94 (0.70-1.27)	41
																	0
	Cabozantinib	259	62 (18-89)	197 (76)	31 (12)	69 (27)	153 (59)	36 (14)	257 (99.3)	134 (51.7)	125 (48.3)	15 (11.8-19.3)	10.8 (10.0-12.5)		NE (21.1-NE)		41
																	1
TiNivo-2	Tivozanib plus nivolumab	171	64 (37-87)	125 (73)	NA	30 (18)	114 (67)	27 (16)	122 (71)	111 (65)	60 (35)	11.8 (11.1-12.6)	5.7 (4.0-7-4)	1.1 (0.84-1.43)	17.7 (15.1-NE)	1.0 (0.68-1.46)	19
																	0.6
	Tivozanib	172	63 (33-82)	134 (78)	NA	31 (18)	113 (66)	28 (16)	122 (71)	105 (61)	67 (39)	12.5 (11.6-13.6)	7.4 (5.6-9.2)		22.1 (15.2-NR)		20
																	0.6

Abbreviations: CI, confidence interval; CR, confidence interval; HR, hazard ratio; ICI; immune checkpoint inhibitor ; ITT, intent to treat; NA, not applicable; ORR, objective response rate; OS, overall survival; PFS, progression free survival.

Figure 1.

Frost plot for progression free survival for intent to treat population (A), IMDC risk groups (B) and treatment lines (C), and overall survival for intent to treat population (D).

Although retrospective and single-arm studies suggested that ICI rechallenge in RCC might have modest activity, phase III randomized clinical trials demonstrated that the efficacy of ICI and VEGF TKI combination was not superior to anti-VEGF TKI monotherapy.^4,5 The combination of pembrolizumab and lenvatinib in the KEYNOTE-146 trial, a single-arm phase 2 study, demonstrated an objective response rate (ORR) of 56% in 104 patients, with only 65% having prior VEGF TKI exposure.⁶ Notably, KEYNOTE-146 was not a randomized trial, and the ORR may primarily reflect lenvatinib’s activity as monotherapy rather than the combination’s synergistic effect. For context, lenvatinib plus everolimus achieved an ORR of 43% in a phase 2 trial where all patients had progressed on VEGF-targeted therapy.⁷ In comparison, the ORRs in the randomized phase 3 trials, CONTACT-03 (41%) and TiNivo-2 (19%), were lower and similar to those observed in the VEGF TKI monotherapy arms. This highlights the variability in response rates, likely influenced by differences in trial design, patient characteristics, and prior treatments. Additionally, the PFS and OS outcomes in CONTACT-03 were superior to TiNivo-2 across both monotherapy and combination arms. Differences in trial design, ICI class (PD-1 vs PD-L1), and anti-VEGF TKI agents (cabozantinib vs tivozanib), as well as baseline and disease characteristics, are important factors to consider. For example, CONTACT-03 included patients with non-clear cell histology and had a higher proportion of IMDC favorable-risk patients. Additionally, in CONTACT-03, both arms received the same dose of cabozantinib, whereas in TiNivo-2, patients in the combination arm started with a lower dose of tivozanib compared to the monotherapy arm, potentially impacting efficacy.

In TiNivo-2, no significant differences in toxicity were observed, with grade ≥3 treatment-related adverse events occurring in 32% of patients in the combination arm and 35% in the monotherapy arm, likely due to the lower tivozanib dose in the combination arm. Conversely, in CONTACT-03, adding atezolizumab increased the rates of serious adverse events by 12% (24% vs 12%) and grade 3-4 treatment-related adverse events by 8% (55% vs 47%), compared to cabozantinib alone, despite no improvement in efficacy. Furthermore, deaths due to adverse events occurred at nearly double the rate in the combination arm. Unlike the durable responses seen in first-line ICI combination studies, the addition of atezolizumab to cabozantinib in CONTACT-03 did not significantly enhance the duration of response.^8–10

There are two pitfalls in the CONTACT-03 trial that should be regarded. First, a PD-L1 inhibitor, atezolizumab, was used as an ICI rechallenge. However, atezolizumab is not an approved treatment option because its combination with bevacizumab did not improve survival compared anti-VEGF TKI alone in the first-line treatment of mRCC.¹¹ Moreover, the role of anti PD-L1 in the treatment of mRCC is questioned due to the limited activity of another PD-L1 inhibitor, avelumab, in combination with axitinib in the first line. The avelumab and axitinib combination demonstrated a better PFS but lacked activity in improving OS.¹² Second, in contrast to the TiNivo-2 trial, CONTACT-03 included patients who were immediately treated with ICI in the preceding line. A pharmacodynamic study of nivolumab indicated that more than 70% of PD-1 molecules on the circulating T-cells were still occupied by the drug even 2 months after the last drug infusion.¹³ Patients with durable responses often show broadened peripheral T- and B-cell repertoires and the development of immunological memory, which may explain why the therapeutic effects of ICIs frequently extend far beyond their pharmacokinetic half-life; recent findings further support a key contribution of memory CD4⁺ T cells to this efficacy.^14,15 However, the TiNivo-2 trial shed light on these two points of skepticism in the CONTACT-03 trial. Nivolumab, a PD-1 inhibitor, is the backbone of the ICI-rechallenge combination, and 30% of all patients enrolled in the study were treated with non-ICI regimens after progression with an ICI-based therapy. Though the TiNivo-2 trial did not demonstrate better OS and PFS in the ICI combination arm compared to anti-VEGF TKI alone.

There are still unanswered questions from these studies. The underrepresentation of patients receiving adjuvant ICI therapy makes it difficult to determine the appropriate strategy for those who progress during adjuvant therapy. Furthermore, these studies do not address treatment planning for patients who experience progression after completing scheduled IO therapy.

Multiple mechanisms contributing to ICI resistance have been identified so far. These mechanisms may stem from the tumor itself, including changes or downregulation of the HLA antigen presentation complex through various pathways, abnormalities in the IFN-gamma/JAK/STAT pathway, RAF, RAS, phosphatidylinositol 3-kinase (PI3K), and PTEN signaling, or from the tumor microenvironment, including T cell exhaustion, heightened activity of regulatory T cells, tumor-associated macrophages, and cancer-associated fibroblasts.¹⁶ Immunotherapy resistance caused by permanent genetic changes is less likely to reverse. However, our understanding of how other resistance mechanisms, such as those related to the tumor microenvironment or epigenetic modifications, might change when immunotherapy is paused or replaced with another treatment remains limited. Studies exploring different strategies to overcome ICI failure have not been successful so far. Adding different classes of ICI agents, such as anti-lymphocyte activation gene 3 (LAG-3) or anti-cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), to nivolumab showed limited activity in patients with prior progression on ICI.^17,18 Similarly, adding ipilimumab to nivolumab in patients who did not achieve an objective response with nivolumab alone improved ORR but offered limited efficacy compared to initial PD-1/CTLA-4 combination therapy.^19–21 Retrospective data suggest that patients who received an alternate therapy between two ICI treatments demonstrated lower response rates during rechallenge in patients with non-small cell lung cancer.²² Similarly, in the TiNivo-2 trial, shorter PFS durations were observed in the cohort where the most recent therapy prior to rechallenge was non-ICI. These shorter PFS durations are likely because patients, after multiple lines of antitumor therapy, often have a heavier tumor burden and poorer physical condition, reducing the efficacy of subsequent ICI treatment.

Further translational research is needed to determine whether ICI rechallenge can effectively restore immune activity after progression. Ongoing trials, mostly using anti-PD-1-based regimens with anti-VEGF TKIs, aim to clarify this issue. Beyond VEGF TKIs, combinations with novel targets are also being studied, including CTLA-4, LAG-3, ILT4, HIF-2α, and histone deacetylase inhibitors. Novel strategies also encompass radioligand therapy and recombinant interleukins and T-cell growth factors.²³

However, it is important to acknowledge certain limitations in our study, primarily the absence of individual-level patient data. This limitation prevents us from conducting more detailed subgroup analyses and personalized treatment recommendations based on specific patient characteristics. Additionally, as only two studies with a relatively small total sample size were included in the meta-analysis, it is not sufficient to identify potential subgroups of patients who might benefit from the treatment.

In conclusion, current evidence indicates that anti–PD-1/PD-L1 rechallenge in combination with a VEGF TKI after progression on ICI does not improve survival outcomes in mRCC, while adding toxicity and potentially compromising the effective TKI dose. This could potentially result in a rise in clinical visits due to adverse events and the need for ICI infusions, ultimately leading to an increase in treatment costs.²⁴ The absence of data on the evolution of the tumor microenvironment following ICI exposure and an ICI break, as well as on intrinsic or acquired resistance mechanisms, further complicates this field of study. It is crucial to first gain a deeper understanding of the tumor and its microenvironment to develop novel approaches to ICI rechallenge.

Author contributions

Hatice Bolek (Conceptualization, Data curation, Investigation, Visualization, Writing—original draft), Sati Coskun Yazgan (Data curation, Visualization), Emre Yekedüz (Conceptualization, Formal analysis, Methodology, Software, Visualization), Bradley A. McGregor (Methodology, Supervision, Writing—review & editing), Rana R. McKay (Methodology, Supervision, Writing—review & editing), Sumanta K. Pal (Methodology, Supervision, Writing—review & editing), Toni K. Choueiri (Methodology, Supervision, Writing—review & editing), and Yüksel Ürün (Conceptualization, Methodology, Supervision, Writing—review & editing)

Supplementary material

Supplementary material is available at The Oncologist online.

Funding

None declared.

Conflicts of interest

H.B., S.C.Y., and E.Y.: declared no conflicts of interest to report. B.A.G.: research funding to institution—Aveo, BMS, Exelixis, Gilead, Pfizer, and SeaGen; consulting—Arcus, Bayer, BMS, Eisai, Exelixis, GenMab, Gilead, Lily, Pfizer, and SeaGen. R.R.K.: consulting and advisory work—Ambrx, Arcus, AstraZeneca, Aveo, Bayer, Blue Earth Diagnostics, Bristol–Myers Squibb, Calithera, Caris, Dendron, Daiichi Sankyo, Eli Lilly, Esiai, Exelixis, Janssen, Merck, Myovant, neomorph, nimbus, Novartis, Pfizer, Sanofi, SeaGen, Sorrento Therapeutics, Telix, Tempus, Institutional Research Support: AstraZeneca, Artera, Bayer, Bristol-Myers Squibb, Exelixis, Oncternal, Tempus. S.K.P.: travel and accommodation fees—CRISPR Therapeutics, MJH Associates, Intrinsiq, Ipsen. T.K.C.: institutional and/or personal paid and/or unpaid support for research, advisory board roles, consultancy, and/or honoraria—Abalytics Oncology, Alkermes, Arcus Bio, AstraZeneca, Aravive, Aveo, Bayer, Bristol Myers-Squibb, Bicycle Therapeutics, Calithera, Circle Pharma, Deciphera Pharmaceuticals, Eisai, EMD Serono, Exelixis, GlaxoSmithKline, Gilead, HiberCell, IQVA, Infinity, Institut Servier, Ipsen, Jansen, Kanaph, Lilly, Merck, Nikang, Neomorph, Nuscan/PrecedeBio, Novartis, Oncohost, Pfizer, Roche, Sanofi/Aventis, Scholar Rock, Surface Oncology, Takeda, Tempest, Up-To-Date, and continuing medical education events (Peerview, OncLive, MJH, CCO, and others) outside the submitted work; institutional patents filed on molecular alterations and immunotherapy response/toxicity and circulating tumor DNA; equity in Tempest, Pionyr, Osel, Precede Bio, CureResponse, InnDura Therapeutics, Primium, and Abalytics; committee roles for the National Comprehensive Cancer Network (Genitourinary Steering Committee), the American Society of Clinical Oncology (Board of Directors), the European Society of Medical Oncology, Academic and Community Cancer Research United (ACCRU), and the Kidney Cancer Coalition (KidneyCAN); and has mentored several non-US citizens on research projects with potential funding (in part) from non-US sources. Y.Ü.: advisory work—Abdi-İbrahim, Astellas, AstraZeneca, Bristol Myers-Squibb, Deva, Eczacıbaşı, Gen ilaç, Gilead, GSK, Janssen, Merck, MSD, Novartis, Pfizer, Roche; travel, honoraria and consultation fees—Abdi-İbrahim, Astellas, Bristol Myers-Squibb, Deva, Eczacıbaşı, Gen İlaç, Gilead, GSK, Janssen, Merck, Novartis, Pfizer, Roche.

Data Availability

Data are available upon reasonable request.