Primary resistance to nivolumab plus ipilimumab therapy in patients with metastatic renal cell carcinoma

Abstract Background Nivolumab plus ipilimumab (NIVO+IPI) is the first‐line treatment for patients with metastatic renal cell carcinoma (mRCC). Approximately 40% of patients achieve a durable response; however, 20% develop primary resistant disease (PRD) to NIVO+IPI, about which little is known in patients with mRCC. Therefore, this investigation aimed to evaluate the clinical implication of PRD in patients with mRCC to select better candidates in whom NIVO+IPI can be initiated as first‐line therapy. Methods This multi‐institutional retrospective cohort study used data collected between August 2015 and January 2023. In total, 120 patients with mRCC treated with NIVO+IPI were eligible. Associations between immune‐related adverse events and progression‐free survival, overall survival (OS), and objective response rate were analyzed. The relationship between other clinical factors and outcomes was also evaluated. Results The median observation period was 16 months (interquartile range, 5–27). The median age at NIVO+IPI initiation was 68 years in the male‐dominant population (n = 86, 71.7%), and most patients had clear cell histology (n = 104, 86.7%). PRD was recorded in 26 (23.4%) of 111 investigated patients during NIVO+IPI therapy. Patients who experienced PRD showed worse OS (hazard ratio: 4.525, 95% confidence interval [CI]: 2.315–8.850, p < 0.001). Multivariable analysis showed that lymph node metastasis (LNM) (odds ratio: 4.274, 95% CI: 1.075–16.949, p = 0.039) was an independent risk factor for PRD. Conclusions PRD was strongly correlated with worse survival rates. LNM was independently associated with PRD in patients with mRCC receiving NIVO+IPI as first‐line therapy and might indicate that a candidate will not benefit from NIVO+IPI.


| INTRODUCTION
The treatment strategy has dramatically changed since immune-oncological (IO) drugs were approved as a treatment option for advanced renal cell carcinoma (RCC). 1 Furthermore, a combination regimen of anti-programmed death-1 antibody (PD-1) with cytotoxic T-lymphocyte antigen-4 (CTLA-4) antibody or vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitor (TKI) has been a standard treatment for patients with metastatic RCC (mRCC) as first-line therapy. 2 Nivolumab (an anti-PD-1 antibody) plus ipilimumab (an anti-CTLA-4 antibody) (NIVO+IPI) is a standard treatment option for patients with mRCC at intermediate and poor risk as per the International Metastatic Renal-Cell Carcinoma Database Consortium (IMDC). 3The 50% overall survival (OS) was >4 years, and 40% of the patients achieved durable responses in CheckMate 214. 4 This survival advantage is superior to VEGFR-TKI monotherapy.In contrast, approximately 20% of patients suffered from a primary resistant condition to NIVO+IPI, which led to a poor prognosis. 5Clinicians are required to identify patients for whom NIVO+IPI might not be beneficial, but most clinicians tend to prolong the NIVO+IP treatment for up to 3 months due to concern for possible pseudoprogression. 6This passive approach may lead to unfavorable consequences for the patients because pseudoprogression is not commonly expected. 7Unfortunately, no predictive factor of primary resistance disease (PRD) currently exists, and clinicians are forced to make treatment decisions without any quantitative information.For patients with PRD, the timely administration of other IO combination regimens might be better to avoid impacting a limited lifetime and unnecessary immune-related adverse events (irAEs).This study aimed to evaluate the risk factors for primary resistance to NIVO+IPI in patients with mRCC.If patients likely to experience PRD by NIVO+IPI treatment could be identified, it could save precious time for the individual patients.
This study also aimed to retrospectively determine the clinical aspects of PRD in patients with mRCC treated with NIVO+IPI using a cohort of 120 patients from multiple institutes throughout Japan.

| Data collection and study cohort
This retrospective multi-institutional study was approved by the review board of all participating institutes (ID: 2245 in the Institutional Review Board of Akita University Graduate School of Medicine).Informed consent was not required for this retrospective study.The inclusion criteria were patients who were treated with NIVO+IPI as a first-line therapy for mRCC.Those with uncontrolled comorbidities were excluded.In total, 120 patients treated with NIVO+IPI for mRCC between August 2015 and January 2023 at 11 institutes were included.The clinical demographics and baseline characteristics of the enrolled patients were analyzed.mRCC was histologically verified in all included patients.

| Definition of primary progressive disease
PRD was defined as a progressive disease diagnosed at the first imaging evaluation or clinical progression within 3 months of treatment initiation.

| Study endpoints
The primary objectives were to compare the OS and progression-free survival (PFS) between patients who showed PRD and those who did not show PRD.The objective response rate was described based on the development of PRD as a secondary objective.The risk factors associated with PRD were screened and confirmed using Cox regression analysis.

| NIVO+IPI treatment
The criteria, dose, and schedule of NIVO+IPI consisted of four cycles of NIVO+IPI and subsequent nivolumab monotherapy until patients developed intolerant irAEs or disease progression was clinically confirmed.The treatment schedule did not strictly follow the drug guidance, and NIVO+IPI was administered at the discretion of each physician.All analyzed patients were categorized as being at intermediate or poor risk according to the IMDC.

| Patient monitoring
The monitoring protocol was not unified and depended on the decision of each physician.We collected data, including demographic information, RCC histology, the number of metastatic sites, body mass index, Eastern Cooperative Oncology Group performance status, complete blood count with differential and platelet, biochemical profile, urinalyses, chest radiography platelets, and irAEs, from the patients' medical records.Neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and systemic immune-inflammation index (calculated by [neutrophil × platelet]/lymphocyte) were investigated as candidate predictive markers for PRD.Lymph node swelling 1 cm or more detected by computed tomography (CT) scan in both regional and extra-regional was diagnosed as lymph node metastasis (LNM).These parameters were re-evaluated during therapy based on the decision of the attending physicians.Toxicity data were collected by investigators and coded according to the National Cancer Institute Common Toxicity Criteria version 2.0.Tumor responses were measured using CT scans and assessed using the Response Evaluation Criteria in Solid Tumors guidelines version 1.1.After starting the IO combination therapy, the admission and assessment schedule was considered carefully by the attending physicians on an individual basis.This report conforms to the STROBE guidelines.

| Statistical analysis
OS was defined as the time interval from the first NIVO+IPI treatment to the date of death from any cause, with censoring at the last follow-up.PFS was defined as the time from the first induction of immunotherapy to disease progression or date of death (whichever occurred first), with censoring at the time of the last follow-up.The database record was closed upon patient death or at the final follow-up.Data were expressed as the median and interquartile range (IQR), and statistical significance was set at p < 0.05.The frequencies of categorical variables were compared using the Pearson chi-square or Fisher's exact test, whichever was appropriate, and the odds ratio (OR) was estimated in proportions.Survival curves were generated using the Kaplan-Meier method and compared using Cox proportional hazard regression analysis.The Cox hazard analysis was also applied to investigate hazard ratio (HR) and 95% confidence interval (CI) in univariate and multivariable analyses.Clinical factors were included in the multivariable analysis if their univariate p-value was <0.1.All data were analyzed using the SPSS version 26.0 statistical software (SPSS Japan Inc.).

| Evaluation of response to treatment (including PRD)
The median follow-up period for the whole cohort was 16 months (IQR, 5-27).All patients were followed up using the analysis.Nine patients were excluded from the analysis owing to the lack of evaluation of the best response to treatment.Twenty-six of 111 patients (23.4%) were evaluated as having PRD based on progression within 3 months of treatment initiation (Table 2) according to imaging modality or clinical course.There were no treatment-related deaths.

| Adverse events
irAEs were observed in 76 patients (63.3%), detailed in Table 3.Of the 76 patients, 56 (50%) had Grade ≥3 severe irAE.Only five patients in the PRD group suffered from severe irAE because their treatment periods were shorter than those of other patients, which resulted in a low rate of irAE.

| Analyses of factors associated with OS
The univariate analysis showed that high C-reactive protein, high neutrophil-to-lymphocyte ratio, and LNM were associated with PRD (Table 1).Multivariable analysis showed that LNM (OR: 4.274, 95% CI: 1.075-16.949,p = 0.039) was an independent risk factor for PRD (Table 4).
In this retrospective, multi-institutional study, 23.4% of patients with mRCC demonstrated primary resistance to NIVO+IPI therapy.OS and PFS were worse in patients diagnosed with PRD.Positive LNM was a risk factor for PRD.To the best of our knowledge, this is the first report to analyze the clinical aspects of PRD in patients with mRCC treated with NIVO+IPI.
Clinical outcomes in patients with mRCC have improved consistently because of the introduction of effective treatment options such as IO combinations.In these IO combinations, NIVO+IPI is shown to be efficacious. 8Approximately 10% of patients accomplished complete disease remission, whereas more than 20% achieved durable responses and presented with a measurable disease without any progression for years. 5These patients were not expected to die of disease progression and thus were thought to be "cured."This result convinced us to use NIVO+IPI with the objective to cure patients in instances of younger age, high tumor burden, and poor response to IO-TKI combinations.However, NIVO+IPI failed to achieve any clinical benefit at any significant rate. 4In a pivotal study, CheckMate 214, 20% of patients did not benefit from NIVO+IPI therapy, similar to the proportion of patients from the present study in whom NIVO+IPI was not efficacious.These patients showed poor survival, as was predicted.Furthermore, some difficulties exist in judging a PD after NIVO+IPI initiation in patients.The first is pseudoprogression, also called tumor flare. 6When an immune-checkpoint inhibitor is used, there is usually a waiting period of at least 2 months to evaluate the treatment efficacy of the IO drug to distinguish it from pseudoprogression. 9 Even this brief delay may be crucial for patients with PRD.The second is hyper or critical progression after NIVO+IPI administration. 10This unexpected reaction did not allow patients a chance to recover from the disease.These two aspects of NIVO+IPI treatment allowed for two distinct treatment strategies: positive selection of patients and avoidance.In the present study, we focused on patient avoidance, thus avoiding a patient in whom NIVO+IPI treatment is less likely to be beneficial.The irresponsible administration of NIVO+IPI without enough provisions tends to be detrimental for patients.Some mechanisms underlying resistance to IO drugs have been investigated. 11However, the actual cause of these mechanisms remains unclear. 2 Efficient IO treatment needs precise function and enough infiltrations by effector T cells in a tumor environment.Tumor resistance to IO therapy is observed when malignant cells impair T cells at different stages of their maturation or effector function. 12Understanding the biological pathway underlying malignant cells' escape from immune surveillance by T-cell suppression will facilitate the selection of better candidates for IO drugs for RCCs. 13,14Primary resistance to NIVO+IPI manifests in approximately 20% of patients with mRCC, leading to a poor prognosis. 5,15Overcoming this unexpected response is essential to clarify the primary resistance mechanisms.
NIVO+IPI demonstrated less effectiveness against LNM in the present study.IO drug has been known for organ-specific effects. 16Anti-PD-1 antibody plus lenvatinib therapy showed less effectiveness against unresectable hepatocellular carcinomas and LNM than against macrovascular tumor thrombi or lung metastases. 17,18A similar result might be expected in thyroid cancer with LN involvement. 19Furthermore, advanced gastric cancer with LNM showed a favorable survival rate under nivolumab treatment. 20These distinct reactions to IO drugs could be due to biological differences among carcinomas and/ or the immune status of infiltrating T cells in LNs.The lymphocyte subset in LN may be associated with a treatment result of the IO drug in patients with LN disease. 21atients with LN diseases using the IO drug might experience a favorable outcome when effector T cells with PD-1(+) CD8(+) dominantly infiltrate the tumors in the LN. 21n contrast, CD4(+) CD39(+) T cells are already divergent from exhausted T cells. 22These exhausted T cells are rarely reprogrammed, even given enough inhibition to the PD-1 pathway. 23Infiltration of Treg cells may also be associated with the negative reaction of LNM. 24Another explanation of the distinct treatment consequence among cancers with LNM is the existence of tertiary LN. 25 Infection-induced cancer, such as gastric or cervical cancer, usually involves tertiary LNs during the chronic inflammatory phase of pre-existing tumorigenesis.Established alternative LN can work similarly to native LN in recruiting effector T cells to the tumor microenvironment. 26This explains the favorable result in patients with gastric cancer with LN involvement.This study had several limitations, including patient and treatment selection biases inherent in the retrospective design.The limited sample size is an additional limitation.However, we performed a Cox multivariable regression analysis to mitigate possible biases associated with confounders, which is the strength of the present study because the existence of LNM did predict PRD independently.
F I G U R E 1 Kaplan-Meier curve of progression-free survival (A) and overall survival (B) in patients with metastatic renal cell carcinoma based on whether they demonstrated primary resistance to nivolumab and ipilimumab treatment.

(A) (B)
T A B 4 Predictive clinical valuable for primary progressive disease in mRCC patients treated with nivolumab plus ipilimumab analyzed by using univariate and multivariate logistic regression models.This retrospective study confirmed that PRD was associated with poor survival and showed that LNM was independently associated with PRD in patients with mRCC treated with NIVO+IPI as first-line therapy.mRCC patients with LNM may benefit from IO combination treatments other than NIVO+IPI.

Risk factor
Patient demographics.