Renal cell carcinoma (RCC) is the 6th most frequently diagnosed cancer in men and the 13th most common cancer in women in Japan [1
]. Although most RCC cases are globally found as an incidental tumor on imaging, survival is highly dependent on the stage at diagnosis, with a 5-year relative survival of only 12% for stage IV metastatic disease [2
]. About one-third of cases are diagnosed as metastatic RCC (mRCC), and 20–50% of patients with RCC who undergo surgery will develop metastatic disease. Initial management for stage IV RCC varies according to prognostic factors [3
]. First-line targeted therapies with less toxicity and high survival benefits have now become the mainstay of treatment for mRCC [4
]. The currently recommended first-line target therapy options in the National Comprehensive Cancer Network guidelines are single-agent tyrosine kinase inhibitors, vascular endothelial growth factor (VEGF) inhibitors, including pazopanib, sunitinib (SUN), axitinib (AXI), and cabozantinib, or everolimus (EVL) and temsirolimus, as mammalian targets of rapamycin pathway (mTOR) inhibitors [2
]. However, only 50% and 20% of the patients with mRCC received second- or third-line treatment after treatment using these drugs, respectively [5
A commonly used, validated model to assess prognosis was developed by the International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) [6
]. The IMDC model classified patients with advanced or mRCC into three groups, namely favorable-, intermediate-, and poor-risk groups, using clinical and laboratory risk factors [6
]. Approximately 75% of patients with advanced or mRCC are in the intermediate- or poor-risk group and have worse oncological outcomes than those in the favorable risk group [6
In the CheckMate 025 trial, NIVO (programmed cell death protein 1: PD-1), which is one of immune checkpoint inhibitors (ICIs), was compared to EVL in patients with clear cell mRCC who previously received anti-VEGF therapy [7
]. Patients who received NIVO had more significantly improved overall survival (OS) than those who were administered EVL (hazard ratio [HR], 0.73; p
= 0.002) [7
]. In the CheckMate 214 trial, combination therapy with nivolumab plus ipilimumab (cytotoxic T-lymphocyte-associated protein 4; CTLA-4) (NIVO+IPI) was compared with SUN in first-line clear cell mRCC treatment [8
]. In patients with intermediate- or poor-risk disease, according to the IMDC model, the 18-month OS rate was 75% with NIVO+IPI and 60% with SUN, and the median OS was not reached with NIVO+IPI versus 26.0 months with SUN (p
< 0.001) [2
]. Additionally, OS benefits were maintained with NIVO+IPI versus SUN in both intermediate- and poor-risk patients after an extended minimum follow-up of 42 months [9
]. Of these, 60 Japanese patients were enrolled in the CheckMate 214 trial (31 and 29 in the NIVO+IPI and sunitinib arms, respectively) [10
]. Although OS was not significantly different between the two groups (HR, 0.56; 95% confidence interval, 0.19–1.59; p
= 0.267) because of the small sample size, Japanese patients treated with NIVO+IPI showed a delayed OS benefit compared with those treated with SUN [10
]. In addition, the treatment for metastatic RCC has dramatically changed. In an open-label phase III trial (KEYNOTE-426), advanced RCC patients who received pembrolizumab plus AXI had a significantly longer OS and PFS and higher objective response rate than those who received SUN only [11
]. In the phase 3 JAVELIN Renal 101 trial, PFS was significantly longer with avelumab plus AXI than with SUN among patients who received these agents as first-line treatments for advanced RCC [7
]. The results of the IMmotion151 trial for untreated metastatic RCC revealed that, in the programmed cell death1-ligand 1 (PD-L1)-positive population, the median PFS was 11.2 months in the atezolizumab plus bevacizumab group versus 7.7 months in the SUN group (p
= 0.022) [12
]. In CheckMate 9ER, nivolumab (NIVO) plus cabozantinib demonstrated superiority over SUN by doubling the PFS time and OS rate and significantly improving OS for advanced RCC [13
]. Based on these results, it can be suggested that combination therapy may have several advantages with oncological outcomes in advanced or metastatic RCC compared with NIVO monotherapy.
Therefore, we conducted a multicenter, retrospective cohort study to evaluate the efficacy and safety of combination NIVO+IPI in patients with advanced or mRCC.
The clinical characteristics of the patients in this study were generally similar to those of the population of Checkmate 214 [8
]. However, the enrolled patients in this study comprised a relatively higher proportion of patients who were classified as IMDC poor risk than those in Checkmate 214 [8
]. In this study, ORR and DCR were 34.3% and 80.0%, respectively, for advanced or mRCC patients who received NIVO+IPI. Additionally, 8.6% of the patients who were administered NIVO+IPI achieved CR. Tomita et al. reported that ORR, DCR, and CR rates in a Japanese population are consistent with the results observed in the global population [10
]. However, they concluded that further follow-up of the Japanese population may show a late clinical benefit of NIVO+IPI as a first-line treatment for mRCC [10
]. Despite the larger population of IMDC poor-risk patients, our results are comparable with these results. Based on the results of Japanese patients in Checkmate 214 and our results, NIVO+IPI as a first-line therapy may have a potential advantage to achieve the treatment effect for advanced or mRCC patients belonging to the intermediate- or poor-risk group, as stratified using the IMDC model.
In the CheckMate 214 trial with a median follow-up period of 25.2 months, NIVO+IPI had significant benefits for OS and PFS [8
]. The 12-month and 18-month OS rates were 80% and 75% in patients with NIVO+IPI and 72% and 60% in those with SUN, respectively (p
< 0.001) [8
]. These results demonstrated long-term survival benefits and durable responses with NIVO+IPI after extended follow-up of greater than 42 months [9
]. OS and ORR benefits were maintained with NIVO+IPI over SUN in intermediate- and poor-risk patients [9
]. Additionally, the PFS had plateaued after 36 months in intermediate- or poor-risk patients who received NIVO+IPI [9
]. For these reasons, patients with NIVO+IPI significantly achieved CR more often than those with SUN (p
< 0.0001) [9
]. In addition, almost half the complete responders experienced a treatment-free interval [9
]. Therefore, response at 6 months after the initiation of NIVO+IPI may be positively associated with long-term OS for mRCC in the intermediate- or poor-risk groups [9
]. However, the Japanese patients treated with NIVO+IPI had a trend toward a later OS benefit than those treated with SUN, even though OS and PFS were similar in the NIVO+IPI and SUN arms [10
]. For the demographic and baseline characteristics, Japanese patients who received prior radiotherapy or had high stage disease were relatively lower compared with the global population in the CheckMate 214 trial [10
]. These differences may have contributed to the difference of oncological outcomes between the Japanese and global patients [10
]. In this study, oncological outcomes, including OS and PFS, were relatively higher than the results of the CheckMate 214 trial. High DCR may be influential in improving oncological outcomes.
Zhang et al. [19
] reported that systemic inflammation plays a crucial role in the development and progression of cancer. NLR and PLR can be easily determined from the peripheral blood count. Several studies have evaluated the role of such inflammatory cell ratios as predictive biomarkers in patients with various solid tumors treated with NIVO [19
]. Whereas high baseline NLR and PLR were found to be associated with treatment failure and increased risk of death, low NLR following NIVO treatment was associated with improved oncological outcomes [19
]. However, the relationship between NLR and PLR and their predictive effects in mRCC patients treated with NIVO+IPI remain unclear. Recently, several meta-analyses have evaluated the utility of PLR as a prognostic factor in cancer patients treated with ICIs [22
]. To our knowledge, this is the first study to evaluate the utility of PLR as a prognostic marker in mRCC patients who received ICI therapy, particularly NIVO+IPI. Our study suggests that a high pretreatment PLR may be associated with poor PFS in mRCC patients administered NIVO+IPI. Previous studies have revealed the utility of NLR as an inflammatory biomarker in mRCC patients [23
]. Although NLR was not significantly associated with PFS in the multivariate analysis in our study, mRCC patients with NLR < 4.6 reported a significantly longer PFS than their counterparts. Therefore, NLR may be considered a prognostic marker in mRCC patients treated with NIVO+IPI. Further long-term studies are required to verify the effectiveness of PLR and NLR as prognostic biomarkers in mRCC patients treated with NIVO+IPI.
TRAEs in patients with advanced or mRCC who underwent NIVO+IPI were consistent with that in previous studies for multiple tumor types [26
], and a relatively lower incidence of grade 3–4 TRAEs was observed than with SUN [8
]. Especially, TRAEs with NIVO+IPI were low grade, and there was a low incidence of grade 3/4 TRAEs compared with targeted therapies, such as SUN, sorafenib, or axitinib [10
]. Although 15 (42.9%) patients with NIVO+IPI experienced grade 3/4 TRAEs, only 4 patients (11.4%) discontinued NIVO+IPI. However, most Japanese patients may have a manageable safety profile for treatment with NIVO+IPI when combined with a steroid.
There are several limitations to our study. First, this was a retrospective study and was conducted using multicenter data. Therefore, this study had an inherent potential for bias, with diagnostic and therapeutic variations among these institutions. Second, a relatively small number of patients were enrolled in this study, and the follow-up period was relatively short. Third, there was no control group of patients who received TKIs and VEGF or mTOR inhibitors for mRCC. Fourth, we could not evaluate the expression levels of PDL-1 and CTLA-4 because this was a multicenter retrospective study. Indeed, lack of PD-L1 expression correlates with worse outcomes with ICI treatment [29
]. However, several randomized studies have demonstrated that patients with PD-L1- positive tumors did not show improved OS and ORR [7
]. Finally, we did not collect the duration of response and when the TRAEs occurred during the treatment of NIVO+IPI.