Abstract
Background/Aim: Patients with advanced renal cell carcinoma (aRCC) treated with immune-oncology (IO) drugs may need to discontinue the treatment when severe immune-related adverse events (irAE) occur; however, the impact of discontinuation on survival remains unknown. Patients and Methods: This is a retrospective multicenter analysis using a database of 183 aRCC patients treated with first-line IO drugs combination. The patients were divided into two groups according to the necessity of discontinuation due to irAEs. The primary endpoint was overall survival (OS). Cox proportional hazard models determined the predictive factors on OS. Results: Among a total of 135 patients who experienced irAE, 38 patients had to discontinue and 52 continued the treatment while treating irAE. When compared to patients who were able to continue treatment, discontinuation was associated with significantly higher rates of IO-IO doublet use, severe irAE (grade ≥3), steroid use, and the occurrence of immune-related pneumonitis (p=0.03, p<0.001, p<0.001, and p=0.02, respectively). The objective response rates were comparable between the two groups (discontinuation 55.6% vs. no discontinuation 56.0%, p=0.7). On univariate analysis, patients who discontinued had a significantly worse OS when compared to those who continued treatment (p=0.02). On the contrary, on multivariate analysis treatment discontinuation was not associated with poor OS (HR=1.1, p=0.9). Conclusion: Treatment discontinuation due to irAE was not associated with poor prognosis in aRCC patients treated with ICI-based combination therapy. Treatment discontinuation may be a reasonable treatment option for well-selected patients, specifically for those who experienced good treatment responses.
Immune-oncology (IO) drug combinations have been established as the standard of care in advanced or metastatic renal cell carcinoma (mRCC) (1). Patients undergoing these types of treatments may experience immune-related adverse events (irAEs). IrAEs can occur at any time during the course of the disease and affect any organ system. Despite rare, grade 5 irAEs have been reported (1-3). Depending on the severity of an irAE, treatment discontinuation may be considered (2). The management of irAEs as well as the necessity to discontinue treatment has been well described (4). However, the potential occurrence of severe irAEs remains a challenge in clinical practice.
Interestingly, several studies have shown that the occurrence of irAEs was associated with better outcomes (5-7) and patients may experience durable response, especially after dual immune checkpoint inhibition with a cytotoxic T-lymphocyte associated protein 4 (CTLA-4) inhibitor and a programmed death-1 inhibitor (PD-1-inhibitor) (5, 8, 9). IO-discontinuation may be reasonable once the patient has experienced a response (10), however, the impact of discontinuation on survival is less well described. In addition, the occurrence of irAEs was shown to be associated with a higher risk for recurrent irAEs if the patient is re-exposed to this type of treatment (11). Therefore, resuming IO therapy after recovery from irAEs remains challenging.
In this study, we evaluated the impact of discontinuation due to irAEs on the survival of mRCC-patients treated with first-line IO combination therapies.
Patients and Methods
Study population. This retrospective study was conducted on the database of 183 consecutive patients with advanced RCC (aRCC) treated at Okayama University hospital and nineteen associated institutions between August 2015 and September 2022. Patients with metastatic or unresectable locally advanced RCC treated with first-line combination therapies using IO drugs were enrolled. Therapies included both IO-IO combinations as well as IO plus vascular endothelial growth factor receptor tyrosine kinase inhibitor (VEGFR-TKI) therapies. This study was approved by the local ethical committee of Okayama University hospital (Registration number 2207-002).
Data on patient’s age, sex, clinical stage (TNM classification), nephrectomy status, metastatic site, the number of metastatic organs, treatment regimen and International Metastatic RCC Database Consortium (IMDC) risk classification were collected. Severity and type of irAEs as well as the use, dose and administration route of corticosteroids were assessed (high-dose steroid was defined as intravenous or oral administration of ≥40 mg per day). The reason for interruption was recorded for patients who experienced irAEs. If additional events beyond irAEs occurred to consider treatment-interruption, the treating physician had to define the most important reason for interruption.
Protocol of IO drugs combination therapies. In this study, a total of four regimens were administrated (ipilimumab plus nivolumab, pembrolizumab plus axitinib, avelumab plus axitinib, and nivolumab plus cabozantinib). Each drug was administrated according to the schedule of pivotal phase III clinical trials (12-15). Dose reductions of IO drugs were not allowed, while decisions on treatment delays were made by the individual clinician depending on the patients’ conditions. Dose modifications of TKIs were allowed.
Clinical endpoints. The primary endpoint was overall survival (OS). OS was defined as the time from initiation of systemic therapy to death from any cause or the last follow up. Objective response rate (ORR) was evaluated according to Response Evaluation Criteria in Solid Tumor (RECIST) version 1.1. All adverse events were evaluated by individual clinicians using Common Terminology Criteria for Adverse Event (CTCAE) version 5.0. The irAE profiles (occurrence date, involved organs, grade, the use of steroids, and the presence of treatment discontinuation) were identified in the clinical records.
Statistical analysis. All statistical analyses were conducted using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan). Survival functions were estimated using the Kaplan–Meier method, and survival was compared using the log-rank test. Continuous variables were analyzed using the Mann–Whitney U-test, and categorical variables were analyzed using Fisher’s exact test. Uni- and multivariate cox hazard models were used to assess the association between potential prognostic factors and OS.
Results
Of 183 patients treated, 135 experienced irAEs; these led to treatment interruptions in 83 patients (INT-patients), 52 patients did not require treatment interruption (CONT-patients) despite experiencing irAEs. Among the 83 INT-patients, 37 patients had more serious reasons to interrupt beyond their irAEs; these patients were excluded from analysis. The remaining 46 INT-patients -whose major reason for interruption was irAEs- included 8 patients who resumed after recovery from irAEs and 38 patients with discontinuation due to irAEs (DISC-patients). The 8 patients who resumed treatment after recovery were excluded because of insufficient information regarding interruption term. The 38 DISC-patients represented the analyzed cohort with discontinuation due to irAEs (Figure 1).
The flow diagram of patient selection. irAE: Immune-related adverse event; PD: progression disease.
The baseline characteristics of both groups are shown in Table I. 74% of DISC-patients and 50% of CONT-patients received IO doublets (p=0.03) and the median follow up was 15 months for DISC-patients versus 8 months for CONT patients (p=0.037). The numbers of the patients receiving second line VEGFR-TKIs was similar in both groups (21%). Grade 3 or higher irAEs were more common in DISC-patients versus CONT-patients (76% and 23%, respectively, p<0.001). Similarly, the use of steroid and high-dose steroid differed between groups (DISC 82% vs. CONT 23%, p<0.001; and DISC 50% vs. CONT 10%, p=0.00262, respectively). The occurrence of immune-related pneumonitis was significantly more frequent in DISC-patients versus CONT-patients (21% vs. 4%, p=0.016). In contrast, immune-related thyroid dysfunction was more common in CONT- versus DISC-patients (11% vs. 35%, p=0.0122) (Table II).
The baseline characteristics of both patient groups.
The profile of immune-related adverse events.
The Kaplan–Meier curve showed that CONT-patients had significantly longer OS than DISC-patients [median OS; DISC not reached (NR), 95% confidence interval (CI)=18-NR months vs. CONT 33, 95%CI=33-NR months, p=0.02] (Figure 2), while no differences were found with respect to objective response rate (DISC 55.6% vs. CONT 56.0%, p=0.7). In patients with ≥3 irAEs, no significant differences in OS were found (p=0.14). On univariable analysis, eastern cooperation oncology group (ECOG) performance status (PS) ≥2 (reference; <2), IMDC poor risk (reference; favorable and intermediate risk), immune-related nephritis and treatment discontinuation were found to be independent prognostic factors for OS on univariate analysis. On multivariate analysis, only ECOG PS ≥2 and the occurrence of immune-related kidney dysfunction remained significantly associated with worse OS (HR=4.3, p=0.0431; HR=34.64, p=0.001, respectively) while treatment discontinuation did not (HR=1.07, p=0.9) (Table III).
Kaplan–Meier plot of overall survival (OS) depending on discontinuation. NA: Not available; CI: confidential interval; CONT: without discontinuation; DISC: with discontinuation.
Univariate and multivariate analysis on overall survival.
Discussion
The introduction of IO-combinations has considerably improved the outcome of patients with mRCC. However, irAEs may occur, requiring rapid management and often the expertise of a specialized, multidisciplinary team (3, 16). Previous reports have shown that the occurrence of irAEs may be associated with better outcomes, in terms of objective response and complete response rates, even if treatment discontinuation is required (3, 8, 10). However, the impact of discontinuation on survival remains unknown (17, 18). This is particularly true for patients receiving high-dose corticosteroids to manage irAEs.
The aim of this study was to compare outcomes of patients with irAEs who did or did not require treatment interruption and/or discontinuation of IO-based combinations in the setting of mRCC. We found that treatment discontinuation occurs more often in patients treated with IO doublets, however, discontinuation was not an independent predictor of survival in our study. One meta-analysis and one phase II study reported on durable responses after discontinuation of treatment for mRCC patients (18, 19). However, the authors did not report whether the reason for discontinuation was related to irAEs. Komiya et al. analyzed the outcome of patients with advanced non-small cell lung cancer treated with PD-1-inhibitors who discontinued the treatment for several reasons including irAE (20). They showed that patients who discontinued due to irAE experienced better survival than those who discontinued for other reasons. In addition, 9 out of 18 patients with discontinuation due to irAE experienced durable response while being off treatment (range; 1 to 309, and median 117 days). Despite these findings, the reason for IO-discontinuation is of paramount importance; the previously addressed study in patients with non-small cell lung cancer patients revealed that the most common reason for discontinuation beyond irAEs was disease progression (20). Therefore, it remains unknown whether discontinuation due to irAEs is associated with improved survival.
The efficacy and the safety of re-exposure to IO drugs after interruption was demonstrated in several studies (11, 21). When treatment alternatives are lacking, resumption of the former agents may be considered. Ravi et al. reported that 23% of patients who resumed IO treatment after interruption achieved an objective response, when compared to 37% in the first-line setting; in this study, IO drugs used in second-line were different from those in first-line in most patients (11). When compared to patients without irAEs in first-line, those who experienced irAEs were more likely to experience irAEs upon re-exposure (41% vs. 4%, respectively). Alawi et al. showed that the 2-year OS rate in the re-treatment- and discontinuation-cohorts were 76% and 66%, respectively. The authors reported that 50% of patients who resumed IO-based treatment experienced irAEs (a different type of irAE in half of patients and the same irAE in one third) (21). Dolladille et al. showed that 28.8% of 452 patients with re-treatment experienced the same irAE, among these, colitis was the most common (22). Similarly, Bylsma et al. reported that in most cases, the type of irAE occurring in first-line and upon re-exposure is the same (23). Although treatment re-exposure was shown to be associated with response, the occurrence of potentially life-threatening irAEs, such as pneumonitis, myocarditis, and neurological events needs to be considered (22, 24, 25).
According to the current guidelines, permanent discontinuation is strongly recommended upon occurrence of grade 4 irAEs (4, 22). In our study, the occurrence of grade ≥3 irAEs was not a significant prognostic factor for OS on multivariate analysis. In patients with grade ≥3 irAEs, there was no significant difference in terms of OS between DISC-patients and CONT-patients. To date, it remains unclear as to whether the occurrence of severe irAEs has an impact on OS. Ikeda et al. reported that severe irAE requiring treatment discontinuation resulted in worse survival (26). Ishihara et al. found that patients with grade <3 (not severe) irAEs survived longer than both those with severe irAEs and those without irAEs (6). One study showed that early interruption due to severe (grade 3 or 4) irAEs leads to a higher rate of durable response than no interruption (67% vs. 37.5%, free of progression in 6 months). Similarly, another study showed that early interruption due to irAEs (18/23 patients experienced grade ≥3 irAEs) had a significant impact on OS when compared to patients without interruption (27). These observations suggest that detection and successful management of irAEs is crucial (2, 16).
Several studies have shown an association between the organ affected by irAE and survival (6, 28). In this study, the occurrence of immune- related nephritis was an independent prognostic factor. It has been reported that the estimated incidence of acute kidney impairment is close to 5.0% of all ICI recipients and it remains unclear which patients are at risk (29). One multicenter cohort study showed that lower baseline estimated glomerular filtration rate (eGFR) was significantly associated with a higher risk of immune related acute kidney impairment on multivariate analysis (30). In our study, the small number of patients with ir-nephritis (n=6, 3.3%) limited the robust analysis to reveal any association.
This study has several limitations. First, the study design was retrospective and included small numbers of patients. Second, multiple treatment regimens were included. Different regimens have different adverse events profiles, especially CTLA-4 inhibitors tend to cause more severe irAEs compared to PD-1 or PD-L1 inhibitors. In our study, more patients in the DISC-cohort had received IO doublets. Third, as VEGFR-TKIs have unique adverse events, it may be difficult to differentiate between TKI and immune-related-AEs in patients receiving IO-TKI combinations; this is particularly true in the case of transaminitis and skin toxicities (16). Fourth, the decision to discontinue treatment was made by the treating physicians in different institutions. This may have led to over- or underestimation of the clinical situation. The rates of the irAEs in this study were actually higher than in another real-world analysis involving 1209 patients (3), therefore the occurrences were possibly overestimated to a certain degree. Despite these limitations, our findings add to the growing body of literature on real world outcomes in patients with mRCC receiving IO-therapies.
Conclusion
In this study, we revealed that treatment discontinuation due to irAE was not associated with poor OS in aRCC patients treated with first-line ICI-based combination therapy. This result may provide benefit for the patients once experienced both irAEs and objective response, especially in the cases which should be considered carefully to resume after discontinuation because of the difficulty in managing irAEs.
Footnotes
Authors’ Contributions
Kensuke Bekku: Conceptualization, writing-original draft, Manuela Schmidinger: Writing-original draft, writing-review and editing, Satoshi Katayama: Data curation, writing-review and editing, Tatsushi Kawada: Data curation, writing-review and editing, Takafumi Yanagisawa: Writing-review and editing, Takehiro Iwata: Data curation, writing-review and editing, Kohei Edamura: Writing-review and editing, Tomoko Kobayashi: Writing-review and editing, Yasuyuki Kobayashi: Writing-review and editing, Shahrokh F. Shariat: Writing-original draft, writing-review and editing, Motoo Araki: Supervision, conceptualization, writing-review and editing.
Conflicts of Interest
Shahrokh F. Shariat received honoraria from Astellas, AstraZeneca, BMS, Ferring, Ipsen, Janssen, MSD, Olympus, Pfizer, Roche, Takeda; had consulting or advisory roles in Astellas, AstraZeneca, BMS, Ferring, Ipsen, Janssen, MSD, Olympus, Pfizer, Pierre Fabre, Roche, Takeda, and was member of speakers bureau for Astellas, Astra Zeneca, Bayer, BMS, Ferring, Ipsen, Janssen, MSD, Olympus, Pfizer, Richard Wolf, Roche, Takeda. Manuela Schmidinger received honoraria for lectures of advisory boards from Ipsen, Exelixis, BMS, MSD, Merck, EUSA, Eisai, Astellas, Janssen. The other Authors declare no conflicts of interest in relation to this study.
- Received October 19, 2023.
- Revision received November 18, 2023.
- Accepted November 20, 2023.
- Copyright © 2024 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
