MSKCC Risk Group Research Articles

Final prespecified overall survival (OS) analysis of CLEAR: 4-year follow-up of lenvatinib plus pembrolizumab (L+P) vs sunitinib (S) in patients (pts) with advanced renal cell carcinoma (aRCC).

4502 Background: In the phase 3 CLEAR trial, L+P showed clinically meaningful and statistically significant benefits in PFS (primary endpoint) and OS, and improved ORR compared with S in 1L aRCC (Motzer NEJM 2021). Here, we report 4-yr follow-up results from the final prespecified OS analysis of CLEAR (data cutoff: 31 Jul 2022). Methods: Treatment-naïve pts (n=1069) who had aRCC with a clear-cell component were randomized (1:1:1) to receive: L 20 mg PO QD + P 200 mg IV Q3W; or L 18 mg + everolimus 5 mg PO QD; or S 50 mg PO QD (4 wks on/2 wks off). Stratification factors were geographic region and MSKCC prognostic risk group. This final prespecified OS analysis was triggered by ~304 death events in 2 arms. OS, PFS, ORR, duration of response (DOR), and PFS on next-line therapy (PFS2) were assessed for L+P and S. PFS, ORR and DOR were assessed per independent review using RECIST v1.1. Nominal P-values are shown. Results: At a median follow-up (IQR) of 49.8 mos (41.4–53.1) for L+P and 49.4 mos (41.6–52.8) for S, 149 and 159 deaths had occurred, respectively. OS benefit with L+P vs S was maintained (HR, 95% CI; 0.79, 0.63–0.99). OS favored L+P vs S across MSKCC risk groups (HR, 95% CI; favorable [fav]: 0.89, 0.53–1.50; intermediate [int]: 0.81, 0.62–1.06; poor: 0.59, 0.31–1.12). PFS benefit of L+P vs S was maintained (HR, 95% CI; 0.47, 0.38–0.57). PFS favored L+P vs S across MSKCC risk groups (HR, 95% CI; fav: 0.46, 0.32–0.67; int: 0.51, 0.40–0.65; poor: 0.18, 0.08–0.42). ORR was greater with L+P (71.3%; complete response [CR], 18.3%) vs S (36.7%; CR, 4.8%) (relative risk, 95% CI; 1.94, 1.67–2.26). Less pts in the L+P arm (181/355, 51.0%) received subsequent anticancer therapies compared with the S arm (246/357, 68.9%); 56 (15.8%) and 195 (54.6%) received PD-1/PD-L1 checkpoint inhibitors, respectively. Analysis of OS adjusted for subsequent therapies will be presented. PFS2 was longer with L+P vs S (43.3 vs 25.9 mos; HR, 95% CI; 0.63, 0.51–0.77). Grade ≥3 treatment-related adverse events occurred in 74.1% and 60.3% pts in the L+P and S arms, respectively. Conclusions: L+P continues to demonstrate clinically meaningful benefit vs S in OS, PFS, ORR, and CR in the 1L treatment of pts with aRCC at 4-yr follow-up, thus supporting the robustness of the primary analysis data from CLEAR. Clinical trial information: NCT02811861 .[Table: see text]

Prognostic model for overall survival that includes the combination of platelet count and neutrophil–lymphocyte ratio within the first six weeks of sunitinib treatment for metastatic renal cell carcinoma

BackgroundThe association between the combination of platelet count and neutrophil–lymphocyte ratio (COP-NLR) at the time of adverse events during sunitinib treatment and prognosis is unclear, and prognostic models combining the prognostic factors of sunitinib have not been well studied. Thus, we developed a prognostic model that includes the COP-NLR to predict the prognosis of patients with metastatic renal cell carcinoma (mRCC) treated with sunitinib.MethodsWe performed a retrospective cohort study of 102 patients treated with sunitinib for mRCC between 2008 and 2020 in three hospitals associated with Showa University, Japan. The primary outcome was overall survival (OS). The collected data included baseline patient characteristics, adverse events, laboratory values, and COP-NLR scores within the first 6 weeks of sunitinib treatment. Prognostic factors of OS were analyzed using the Cox proportional hazards model. The integer score was derived from the beta-coefficient (β) of these factors and was divided into three groups. The survival curves were visualized using the Kaplan–Meier method and estimated using a log-rank test.ResultsThe median OS was 32.3 months. Multivariable analysis showed that the number of metastatic sites, Memorial Sloan Kettering Cancer Center risk group, number of metastases, non-hypertension, modified Glasgow Prognostic Score, and 6-week COP-NLR were significantly associated with OS. A higher 6-week COP-NLR was significantly associated with a shorter OS (p < 0.001). The β values of the five factors for OS were scored (non-hypertension, mGPS, and 6-week COP-NLR = 1 point; number of metastatic sites = 2 points; MSKCC risk group = 3 points) and patients divided into three groups (≤ 1, 2–3, and ≥ 4). The low-risk (≤ 1) group had significantly longer OS than the high-risk (≥ 4) group (median OS: 99.0 vs. 6.2 months, p < 0.001).ConclusionsThis study showed that the COP-NLR within the first 6 weeks of sunitinib treatment had a greater impact on OS than the COP-NLR at the start of sunitinib treatment. The developed prognostic model for OS, including the 6-week COP-NLR, will be useful in decision-making to continue sunitinib in the early treatment stage of patients with mRCC.

Logical Imputation to Optimize Prognostic Risk Classification in Metastatic Renal Cell Cancer

BACKGROUND: Application of the MSKCC and IMDC models is recommended for prognostication in metastatic renal cell cancer (mRCC). Patient classification in MSKCC and IMDC risk groups in real-world observational studies is often hampered by missing data on required pre-treatment characteristics. OBJECTIVE: To evaluate the effect of application of easy-to-use logical, or deductive, imputation on MSKCC and IMDC risk classification in an observational study setting. PATIENTS AND METHODS: We used data on 713 mRCC patients with first-line sunitinib treatment from our observational European multi-centre study EuroTARGET. Pre-treatment characteristics and follow-up were derived from medical files. Hospital-specific cut-off values for laboratory measurements were requested. The effect of logical imputation of missing data and consensus versus hospital-specific cut-off values on patient classification and the subsequent models’ predictive performance for progression-free and overall survival (OS) was evaluated. RESULTS: 45% of the patients had missing data for≥1 pre-treatment characteristic for either model. Still, 72% of all patients could be unambiguously classified using logical imputation. Use of consensus instead of hospital-specific cut-offs led to a shift in risk group for 12% and 7% of patients for the MSKCC and IMDC model, respectively. Using logical imputation or other cut-offs did not influence the models’ predictive performance. These were in line with previous reports (c-statistic ∼0.64 for OS) CONCLUSIONS: Logical imputation leads to a substantial increase in the proportion of patients that can be correctly classified into poor and intermediate MSKCC and IMDC risk groups in observational studies and its use in the field should be advocated.

Cross-trial validation of molecular subtypes in patients with metastatic clear cell renal cell carcinoma (RCC): The JAVELIN Renal 101 experience.

4531 Background: Vascular endothelial growth factor (VEGF) and immune checkpoint inhibitors (IO) combinations are a standard in mRCC. Molecular clusters of patients have been identified and correlated with outcomes in the phase 3 IMmotion151 (IM151) trial of atezolizumab + bevacizumab (IO+VEGF) vs. sunitinib (Sun) (Motzer, Cancer Cell 2020 &amp; JAMA Oncol 2021). Avelumab+axitinib (AA) is an approved IO+VEGF combination in mRCC. This work aims to evaluate these clusters in patients from the phase 3 JAVELIN Renal 101 (JR101; NCT02684006 ) trial of AA vs. Sun. Methods: Bulk RNA-sequencing of primary and metastatic samples and clinical data (data cutoff: 28 January 2019) from JR101 were obtained. A random forest model designed to predict molecular clusters based on transcriptomic data was trained on the IM151 dataset. Using this model, patients from JR101 study were categorized into previously defined molecular subgroups. We then evaluated treatment outcomes including progression-free survival (PFS), objective response rate (ORR), and overall survival (OS) from JR101 in relation to molecular subgroups. Results: The proportion of patients in each molecular subtype and across MSKCC risk groups were largely comparable between the 2 trials (accuracy: 81.6%; p=0.2). AA was generally superior to Sun. for PFS and ORR across all molecular subsets, including angiogenic and immune-based clusters (Table). Combining immune and/or cell cycle-enriched clusters 4+5 resulted in improved PFS (HR: 0.65; 95% CI: 0.44-0.97) for AA vs. Sun. Conclusions: We were able to largely validate the molecular clusters classification and some of the associations with survival outcomes from IM151 in the JR101 clinical trial cohort. Biomarkers of specific VEGF+IO combinations in mRCC should be prospectively validated in randomized trials. [Table: see text]

Phase II randomized double blind trial of axitinib (Axi) +/- PF-04518600, an OX40 antibody (PFOX) after PD1/PDL1 antibody (IO) therapy (Tx) in metastatic renal cell carcinoma (mRCC).

4529 Background: Immune checkpoint blockade has revolutionized mRCC Tx, but primary and acquired resistance continues to result in poor patient outcomes. OX40 (CD-134) mediates IO resistance. Co-stimulatory OX40 (CD-134) activates exhausted T-cells. OX40 activation in dendritic cells increases the proliferation, effector function, and survival of T cells. PFOX is an agonist for OX40. We hypothesized that PFOX + the VEGFR inhibitor Axi would improve outcomes vs. Axi in patients (pts) with mRCC after IO Tx. Methods: Pts with predominantly clear cell mRCC were stratified for MSKCC risk groups then randomized 1:1 to Axi 5mg po bid plus PFOX 0.3mg/kg iv (Arm 1) or placebo (PL) iv (Arm 2) on Day 1 of a 2-week cycle. The primary endpoint was progression free survival (PFS); secondary endpoints included overall survival (OS), objective response rate (ORR), duration of response (DOR) per RECIST v1.1, and safety/tolerability. A prespecified interim analysis (IA) tested PFS at a 1-sided P &lt; 0.02 when ≥ 33 events were observed. Results: Between February 2018 and October 2021 a total of 59 pts were randomly assigned and treated with Axi+PFOX (N = 29) or Axi+PL (N = 30). Pt and disease characteristics are summarized in the table. As of October 2021, 38 PFS events had occurred, 19 on each arm. The IA rejected the hypothesis of added efficacy for PFOX with a p of 0.0089. Subsequently the study was closed to new accrual. At a median follow up of 13.4 mo, median PFS was 13.1 (6-15.8) months (mo) for Arm 1 and 8.5 (5.5-11) mo for Arm 2 (HR = 0.85 [95% CI: 0.45-1.60] p = 0.61). After adjusting by MSKCC risk group and prior lines of Tx, HR = 0.74 [95% CI: 0.38-1.46] p = 0.39. Median OS was not reached (adjusted HR = 0.71 [95% CI: 0.24-2.12] p = 0.54). ORR Arm 1: 31% PR, 52% SD, 14% PD, and Arm 2: 37% PR, 50% SD, 13% PD. Median DOR 9.1 (3.3-23.9) mo for Arm 1, and 7.5 (1.8-32.7) mo. for Arm 2. Rates of any grade Tx related adverse events (TRAEs; 93% vs 100%), including grade 3 or 4 TRAEs (66% vs 47%), in Arm 1 and Arm 2, respectively. 4 pts discontinued Tx due to TRAE, 3 in Arm 1 (1 grade 3 hypertension, 1 grade 2 stroke, 1 grade 3 bullous dermatitis) and 1 in Arm 2 (grade 4 ALT elevation). The most common TRAEs were diarrhea 52%, hypertension 52%, fatigue 41%, nausea 41% for Arm 1 and hypertension 67%, diarrhea 53%, fatigue 50% for Arm 2. Conclusions: In IO-pretreated mRCC pts, Axi + PFOX did not improve outcomes compared to Axi alone. Clinical trial information: NCT03092856. [Table: see text]

Metastatic renal cell carcinoma to the brain: optimizing patient selection for gamma knife radiosurgery.

The effects of single-fraction gamma knife radiosurgery (sf-GKRS) on patients with renal cell carcinoma (RCC) brain metastases (BM) in the era of targeted agents (TA) and immune checkpoint inhibitors (ICI) are insufficiently studied. Clear cell metastatic RCC patients treated with sf-GKRS due to BM in 2005-2014 at three European centres were retrospectively analysed (n = 43). Median follow-up was 56 months. Ninety-five percent had prior nephrectomy, 53% synchronous metastasis and 86% extracranial disease at first sf-GKRS. Karnofsky performance status (KPS) ranged from 60 to 100%. Outcome measures were overall survival (OS), local control (LC) and adverse radiation effects (ARE). One hundred and ninety-four targets were irradiated. The median number of targets at first sf-GKRS was two. The median prescription dose was 22.0 Gy. Thirty-seven percent had repeated sf-GKRS. Eighty-eight percent received TA. LC rates at 12 and 18 months were 97% and 90%. Median OS from the first sf-GKRS was 15.7 months. Low serum albumin (HR for death 5.3), corticosteroid use pre-sf-GKRS (HR for death 5.8) and KPS < 80 (HR for death 9.1) were independently associated with worse OS. No further prognostic information was gleaned from MSKCC risk group, synchronous metastasis, age, number of BM or extracranial metastases. Other prognostic scores for BM radiosurgery, including DS-GPA, renal-GPA, LLV-SIR and CITV-SIR, again, did not add further prognostic value. ARE were seldom symptomatic and were associated with tumour volume, 10-Gy volume and pre-treatment perifocal oedema. ARE were less common among patients treated with TA within 1 month of sf-GKRS. We identified albumin, corticosteroid use and KPS as independent prognostic factors for sf-GKRS of clear cell RCC BM. Studies focusing on the prognostic significance of albumin in sf-GKRS are rare. Further studies with a larger number of patients are warranted to confirm the above analytical outcome. Also, in keeping with previous studies, our data showed optimal rates of local tumour control and limited toxicity post radiosurgery, rendering GKRS the tool of choice in the management of RCC BM.

Clinical Outcomes by Nephrectomy Status In METEOR, A Randomized Phase 3 Trial of Cabozantinib Versus Everolimus in Patients with Advanced Renal Cell Carcinoma

Background: We investigated outcomes with cabozantinib versus everolimus in patients with advanced renal cell carcinoma (RCC) with or without prior nephrectomy in the phase 3 METEOR trial (NCT01865747). Methods: Patients (N = 658) with advanced clear cell RCC and prior treatment with≥1 VEGFR tyrosine kinase inhibitor (TKI) were randomized to cabozantinib 60 mg/day or everolimus 10 mg/day. Pre-specified subgroup analyses of progression-free survival (PFS), overall survival (OS), and objective response rate (ORR) were conducted by prior nephrectomy status. Response was assessed by independent radiology committee. Results: Most enrolled patients (85%) had prior nephrectomy. Baseline prognostic factors (e.g. MSKCC risk group) were less favorable for patients without prior nephrectomy. Cabozantinib improved outcomes versus everolimus in the subgroups with and without nephrectomy—hazard ratios (95% CIs) of 0.51 (0.41–0.64) and 0.51 (0.30–0.86), respectively, for PFS, and 0.66 (0.52–0.84) and 0.75 (0.44–1.27), respectively, for OS. Median OS was numerically longer in patients with versus those without prior nephrectomy in both treatment arms. ORR for cabozantinib versus everolimus was 17% versus 4% for the prior nephrectomy subgroup and 21% versus 2% for the subgroup without prior nephrectomy. Among evaluable patients without prior nephrectomy, reductions of renal target lesions occurred in 94% (16/17) of patients in the cabozantinib arm versus 44% (8/18) in the everolimus arm. The safety profiles of both subgroups were generally consistent with that of the overall study population. Conclusion: Cabozantinib improved PFS, ORR, and OS compared with everolimus in patients with advanced RCC irrespective of nephrectomy status.

Molecular characteristics of renal cell carcinoma (RCC) risk groups from JAVELIN Renal 101.

744 Background: The phase 3 JAVELIN Renal 101 trial in patients (pts) with advanced RCC demonstrated a progression-free survival (PFS) benefit and higher objective response rate (ORR) with avelumab + axitinib (A+Ax) vs sunitinib (S) (Motzer NEJM 2019). PFS and ORR favored A+Ax in all MSKCC risk groups, but median PFS varied. Here, we report results from analyses of baseline tumor samples to define molecular characteristics underlying risk group classifications. Methods: Nephrectomy or tumor samples from pts enrolled in the study were characterized by immunohistochemistry (CD8 and PD-L1), whole exome sequencing, or gene expression profiling (n = 705–850). Gene expression signatures (GES), pathway activation status, and mutational profiles were examined in relation to MSKCC risk groups. Results: Of the 886 total pts enrolled, 23%, 66%, and 11% had favorable (F), intermediate (I), or poor (P) MSKCC risk factors at baseline. In the F, I, and P groups, the ORR was 66%/38%, 50%/24%, and 31%/9%; median PFS was NR/16.7 mo, 13.3 mo/7.9 mo, and 5.6 mo/2.8 mo for the A+Ax/S arms, respectively. Neither the presence of PD-L1+ immune cells nor CD8+ cells differentiated the subgroups; however, the presence of PD-L1+ tumor cells was highest in the P group (p=0.0159). When compared to the I and P groups, the F group was enriched for NOTCH2 mutations (p=0.0002), displayed high FLT1 expression (p=0.007), and showed a trend favoring angiogenesis GES (JAVELIN Renal 101 and IMmotion150). The I group displayed few distinguishing characteristics (low neutrophil GES [p=0.02] and elevated homeobox gene expression [p=0.00052]). Relative to the F group, the P group showed higher cell cycle gene expression (p=0.0057) and PTEN mutation frequency, wild-type NOTCH2 genotype, elevated IMmotion150 Myeloid inflamed GES (p=0.0024), and macrophage-specific GES (p=0.0327), as well as GES for TH2 (p=0.0002), hypoxia (p=0.0199), glycolysis (p=0.0066), and the lowest expression of a dendritic cell GES (p=0.0209). Conclusions: In advanced RCC patients, biological differences within each MSKCC risk category may impact response to treatment and may help explain why these groups perform differently. Clinical trial information: NCT02684006.

1269P - Results from a randomised phase I/II trial evaluating the safety and antitumour activity of anti-PD-1 (MEDI0680)/anti-PD-L1 (durvalumab) vs anti-PD-1 (nivolumab) alone in metastatic clear cell renal cell carcinoma (ccRCC)

BackgroundMEDI0680 is a humanised IgG4κ anti-programmed cell death-1 (PD-1) mAb. We hypothesised that simultaneous blockade of PD-1:PD-L1/PD-L2 with MEDI0680 (M) and PD-1:PD-L1/CD80 with anti-PD-L1 mAb durvalumab (D) would improve efficacy vs blockade of the PD-1:PD-L1/PD-L2 pathway (with nivolumab; N) alone. M+D was well tolerated in the dose-escalation phase of a Phase I/II study in pts with advanced solid tumours, with an ORR of 33% (10/30; including 3/4 RCC pts). In the Phase II portion of the study, we compared M+D to N in a dose-expansion cohort of pretreated, immunotherapy (IO)-naïve pts with metastatic ccRCC. MethodsEligible pts had received 1–3 prior therapy lines, no prior IO exposure and ≥1 measurable lesion. They were randomised 2:1 (stratified by MSKCC risk group and PD-L1 expression) to M 20mg/kg IV + D 750mg Q2W or N 240mg IV Q2W until unacceptable toxicity or disease progression, for ≤2 years. Endpoints included investigator-assessed ORR by RECIST v1.1 (primary endpoint), PFS and safety (secondary). Sample size was ∼60 to detect a difference of 26.0% (ie, ORR = 47.5%, assuming ORR of 21.5% for N) with 76% power at a 1-sided significance level of 0.10. ResultsBy Feb 24, 2019, 63 pts were randomised. Baseline pt/disease characteristics were generally well balanced, but more pts on N had favourable MSKCC risk (7/21; 33.3%) vs M+D (10/42; 23.8%). ORR was 14.3% (6/42; 2 CR, 4 PR; plus 2 unconfirmed PR) vs 19.0% (4/21; 4 PR, 0 unconfirmed) for M+D and N, respectively. There was no difference between arms in ORR by PD-L1 expression (<1% vs≥1%). All responses are ongoing. Median PFS was 3.6 months in both arms. Grade 3/4 treatment-related AEs (TRAEs) occurred in 26% on M+D (including 1 case of autoimmune encephalitis) and 19% on N. On M+D, 12% had TRAEs leading to treatment discontinuation, including colitis or diarrhoea (n=3) and increased ALT/transaminases (n=2), vs 5% on N (pancreatitis and increased lipase and amylase in 1pt). There were no Grade 5 TRAEs. ConclusionsEfficacy was similar with combined M+D and N monotherapy in pts with TKI-pretreated, IO-naïve, metastatic ccRCC, but more pts discontinued M+D due to TRAEs. Clinical trial identificationNCT02118337. Editorial acknowledgementAaron Korpal, PhD, of Cirrus Communications (Macclesfield, UK), an Ashfield company, funded by AstraZeneca. Legal entity responsible for the studyAstraZeneca. FundingAstraZeneca. DisclosureM.H. Voss: Research grant / Funding (institution): BMS; Research grant / Funding (institution): Genentech; Honoraria (self): Eisai; Honoraria (self): Exelixis; Honoraria (self): Pfizer; Honoraria (self): Novartis; Honoraria (self): Calithera; Honoraria (self): Corvus. A.A. Azad: Honoraria (self), Advisory / Consultancy: Janssen; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution), Non-remunerated activity/ies: Astellas; Honoraria (self), Advisory / Consultancy: Novartis; Research grant / Funding (institution), Non-remunerated activity/ies: Merck Serono; Honoraria (self), Advisory / Consultancy: Tolmar; Honoraria (self), Advisory / Consultancy, Non-remunerated activity/ies: Amgen; Honoraria (self), Advisory / Consultancy: Pfizer; Honoraria (self): Bayer; Honoraria (self), Advisory / Consultancy: Telix Pharmaceuticals; Honoraria (self), Advisory / Consultancy: Bristol-Myers Squibb; Honoraria (self), Advisory / Consultancy: Sanofi. A.R. Hansen: Advisory / Consultancy, Research grant / Funding (institution): Genentech/Roche; Advisory / Consultancy, Research grant / Funding (institution): Merck; Advisory / Consultancy, Research grant / Funding (institution): GSK; Advisory / Consultancy, Research grant / Funding (institution): Bristol-Myers Squibb; Advisory / Consultancy, Research grant / Funding (institution): Novartis; Research grant / Funding (institution): Boston Biomedical; Advisory / Consultancy, Research grant / Funding (institution): Boehringer Ingelheim; Advisory / Consultancy, Research grant / Funding (institution): AstraZeneca. J.E. Gray: Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution): AstraZeneca; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution): Genentech; Honoraria (self), Advisory / Consultancy: Celgene; Honoraria (self), Advisory / Consultancy: Takeda; Honoraria (self), Advisory / Consultancy: Janssen; Honoraria (self), Advisory / Consultancy: Eli Lilly; Honoraria (self), Advisory / Consultancy: Triptych Health Partners; Research grant / Funding (institution): Array; Research grant / Funding (institution): Merck; Research grant / Funding (institution): Epic Sciences; Research grant / Funding (institution): BMS; Research grant / Funding (institution): BI; Research grant / Funding (institution): Trovagene; Research grant / Funding (institution): Loxo; Research grant / Funding (institution): Blueprint; Research grant / Funding (institution): Novartis. I. Achour: Shareholder / Stockholder / Stock options, Full / Part-time employment: AstraZeneca. H. Hu: Full / Part-time employment: AstraZeneca. L. Lewis: Travel / Accommodation / Expenses, Full / Part-time employment: AstraZeneca. F.L. Walcott: Shareholder / Stockholder / Stock options, Full / Part-time employment: AstraZeneca. S.F. Oosting: Research grant / Funding (institution): Celldex; Research grant / Funding (institution): Novartis. All other authors have declared no conflicts of interest.

Cytoreductive nephrectomy (CN) in metastatic renal cancer (mRCC): Update on Carmena trial with focus on intermediate IMDC-risk population.

4508 Background: Carmena was a randomized phase III trial, testing the benefit of CN followed by sunitinib (arm A) vs sunitinib alone (arm B), with stratification by MSKCC risk groups in 450 mRCC patients. Based on this trial, CN is not anymore recommended in mRCC (NEJM, Mejean et al, 2018). However there are questions about which patients could still benefit from CN, especially in intermediate risk group. In the present study, we investigated different subgroups from the Carmena trial to answer these questions. Methods: Carmena trial was initially stratified according to MSKCC risk groups. For the purpose of this analysis, we reclassified the patients based on IMDC risk groups. We also analyzed patients with one metastatic site vs more than one, as well as patients with secondary nephrectomy in arm B. Overall survival (OS) was the primary endpoint. Results: With a updated median FU of 61.5 months (mo), the median OS by ITT analysis was 15.6 vs 19.8 mo in arm A and B respectively stratified on MSKCC (HR 0.933 ; 95% CI [0.76- 1.15]) / stratified on IMDC (HR 0.957 ; 95% CI [0.78- 1.18]). Using IMDC risk group factors, 58.6% patients were intermediate and 41.4 % were poor risk. When looking at intermediate risk group only, 48.1% had only one risk factor (interval between diagnosis and treatment &lt; 1y), with a median OS of 30.5 and 25.2 mo in arm A and B respectively (HR 1.24 [0.81 – 1.90]). By contrast, 51.9 % had two risk factors (mostly low hemoglobin, high corrected calcium or neutrophils), with a median OS of 16.6 and 31.2 mo in arm A and B respectively (HR 0.61 [0.41 – 0.91] p = 0.015). Regarding number of metastatic sites, 33% had only one metastatic site. Median OS was 23.6 and 22.7 mo in arm A and B respectively (HR 1.08 [0.75 – 1.57]. Finally, 40 patients had a secondary nephrectomy in arm B, with median OS of 48.5 mo [CI 95%: 27.9-64.4] vs 15.7 mo [CI 95%: 13.3-20.5] in patients who never had surgery. Conclusions: With longer FU, Carmena trial confirms that CN is not superior to sunitinib alone in ITT population, both with MSKCC and IMDC risk groups. However for patients with only one IMDC risk factor, CN might be beneficial. Number of metastatic site is not helpful to define good candidates for surgery. Finally, patients with secondary nephrectomy have very long OS, supporting this strategy. Clinical trial information: NCT00930033.

Analysis of overall survival (OS) based on early tumor shrinkage in the phase III METEOR study of cabozantinib (cabo) versus everolimus (eve) in advanced renal cell carcinoma (RCC).

550 Background: In the phase 3 METEOR study (NCT01865747), cabo improved OS (median 21.4 vs 16.5 mo; HR, 0.66; 95% CI, 0.53–0.83), progression-free survival and objective response rate compared with eve in patients (pts) with previously treated advanced RCC (Choueiri 2016). Retrospective studies have shown that early tumour shrinkage (eTS), based on target lesion reduction from baseline to first post-baseline scan, has predictive value for targeted therapies in RCC (Grunwald 2015; Grunwald 2016); here we evaluate its impact on OS in METEOR. Methods: In total, 658 pts were randomized 1:1 to receive cabo (60 mg qd) or eve (10 mg qd), stratified by MSKCC risk group and number of prior VEGFR TKIs. Target lesion size was assessed by independent radiology review using CT/MRI scans at baseline, every 8 wk for the first 12 mo and every 12 wk thereafter. Median OS was estimated for pts with ≥30% eTS, any eTS or no eTS at first post-baseline scan (week 8); data cutoff, 2 October 2016 (Motzer 2018). Results: Median follow-up was 28 mo (IQR 25, 30). Median (range) time to objective response was 1.91 (1.6, 11.0) mo with cabo and 2.14 (1.9, 9.2) mo with eve, and corresponded to the time to the first post-baseline scan. A greater proportion of pts had ≥30% eTS with cabo (20%) than with eve (5%) and the rate of any eTS was higher in the cabo arm (73%) than with eve (47%; Table). Median OS with cabo vs eve for pts with ≥30% eTS was not reached (NR; 95% CI, 23.7–NR) vs 10.2 mo (95% CI, 3.9–NE), respectively (stratified HR, 0.45; 95% CI, 0.21–0.95; p&lt;0.05). Median OS with cabo vs eve for pts with any eTS was 23.7 (95% CI, 21.7–27.7) vs 17.3 mo (95% CI, 15.4–20.8), respectively; (stratified HR, 0.62; 95% CI, 0.48–0.80; p&lt;0.05). OS was similar for cabo and eve for pts with no eTS. Conclusions: Cabo demonstrated a higher rate and greater magnitude of eTS at first post-baseline scan compared with eve, and eTS was associated with prolonged OS in pts treated with cabo. Clinical trial information: NCT01865747. [Table: see text]

Subgroup analysis from JAVELIN Renal 101: Outcomes for avelumab plus axitinib (A + Ax) versus sunitinib (S) in advanced renal cell carcinoma (aRCC).

544 Background: In the ongoing phase 3 JAVELIN Renal 101 trial, progression-free survival (PFS) was longer (median, 13.8 vs 8.4 mo; hazard ratio, 0.69; p=0.0001) and the objective response rate (ORR) was higher (51% vs 26%) with A + Ax vs S in patients with previously untreated aRCC. Here we report outcomes from an analysis of several prespecified subgroups. Methods: Patients were randomized 1:1 to receive A (10 mg/kg) IV every 2 weeks + Ax (5 mg) PO twice daily or S (50 mg) PO once daily for 4 wk (6-wk cycle). Primary and key secondary endpoints were PFS per independent review committee (IRC; RECIST v1.1) and OS in patients with PD-L1+ tumors (≥1% of immune cells) and in patients irrespective of PD-L1 expression; other secondary endpoints included OR per IRC (RECIST v1.1). Results: A total of 886 patients were randomized; 560 (63%) had PD-L1+ tumors. At data cut-off (Jun 2018), median follow-up was 12.0 vs 11.5 mo for A + Ax vs S groups. The table shows PFS and ORR by MSKCC and IMDC risk groups (F, favorable; I, intermediate; P, poor) and PD-L1 subgroup. Similar results for prognostic risk were seen in patients with PD-L1+ tumors. Outcome data (including PFS2) for additional clinical subgroups by baseline demographics and features will be presented. Clinical trial information: NCT02684006. Conclusions: A + Ax demonstrated PFS and OR benefit across all prognostic risk groups and PD-L1 subgroups vs S in aRCC.[Table: see text]

PTEN Expression, Not Mutation Status in TSC1, TSC2, or mTOR, Correlates with the Outcome on Everolimus in Patients with Renal Cell Carcinoma Treated on the Randomized RECORD-3 Trial.

Genomic alterations in key components of PI3K/mTOR pathway have been proposed as candidate predictive markers for rapalog therapy in renal cell carcinoma (RCC). We tested this hypothesis in patients from a randomized phase II trial of everolimus versus sunitinib. Archival specimens collected at baseline were analyzed with targeted next-generation sequencing (NGS). Focus of interest were alterations in key PI3K pathway components. PTEN expression was assessed by IHC. Association between molecular findings and treatment outcomes was investigated; same associations were tested for 2 everolimus-treated trial cohorts in gastric and hepatocellular carcinoma (HCC). Among 184 everolimus-treated patients with RCC with NGS data, mutation rates in genes of interest were 6% (TSC1), 4.4% (TSC2), and 8.2% (mTOR); 44% harbored alterations in ≥1 PI3K pathway component. For subjects with presence versus absence of mutations in TSC1, TSC2, or mTOR progression-free survival (PFS) neither differed on univariate analysis (HR, 1.0; P = 0.895) nor on multivariate testing stratified by MSKCC risk group and other established prognostic factors (HR, 1.1; P = 0.806). Everolimus-treated patients with retained (n = 50) versus lost (n = 50) PTEN IHC expression had median PFS of 5.3 months versus 10.5 months (HR, 2.5; P < 0.001). Such differences were not seen with sunitinib (10.9 months vs. 10.3 months; HR, 0.8; P = 0.475). Molecular findings did not correlate with outcomes in gastric and HCC cohorts. Association between mutation status for TSC1/TSC2/mTOR and therapeutic outcome on everolimus was not confirmed. Clinically meaningful differences in PFS were seen based on PTEN expression by IHC, lost in >50% of patients.

Baseline Edmonton Symptom Assessment System and survival in metastatic renal cell carcinoma.

Baseline symptom burden as measured using the Edmonton Symptom Assessment System (esas), a patient-reported, validated, and reliable tool measuring symptom severity in 9 separate domains, might yield prognostic information in patients receiving treatment for metastatic renal cell carcinoma (mrcc) and might add to the existing prognostic models. In this retrospective single-centre cohort study, we included patients receiving first-line sunitinib therapy for mrcc between 2008 and 2012. Baseline variables included information relevant to the pre-existing prognostic models and pre-treatment esas summation scores (added together across all 9 domains), with higher scores representing greater symptom burden. We used Kaplan-Meier curves and Cox regression modelling to determine if symptom burden can provide prognostic information with respect to overall survival. We identified 68 patients receiving first-line therapy for mrcc. Most had intermediate- or poor-risk disease based on both the Memorial Sloan Kettering Cancer Center (mskcc) and the International Metastatic Renal Cell Carcinoma Database Consortium (imdc) models. The median baseline esas summation score was 16 (range: 6-57). In univariable analysis, the hazard ratio for overall survival was 1.270 (p = 0.0047) per 10-unit increase in summation esas. In multivariable analysis, the hazard ratio was 1.208 (p = 0.0362) when controlling for mskcc risk group and 1.240 (p = 0.019) when controlling for imdc risk group. Baseline symptom burden as measured by esas score appears to provide prognostic information for survival in patients with mrcc. Those results should encourage the investigation of patient-reported symptom scales as potential prognostic indicators for patients with advanced cancer.

CARMENA: Cytoreductive nephrectomy followed by sunitinib versus sunitinib alone in metastatic renal cell carcinoma—Results of a phase III noninferiority trial.

LBA3 Background: Cytoreductive nephrectomy (CN) has been the standard of care in mRCC in the past twenty years, supported by randomized and large retrospective studies. However the efficacy of targeted therapies has challenged this standard. CARMENA was designed to answer the question of whether upfront CN should continue to be performed before sunitinib. Methods: CARMENA was a randomized phase III trial. Patients (pts) with synchronous mRCC, amenable to CN, were enrolled after confirmation of clear cell histology on biopsy if PS 0-1, absence of symptomatic brain metastasis, acceptable organ function and eligible for sunitinib therapy. Pts were randomized 1:1 to either CN followed by sunitinib (arm A) or sunitinib alone (arm B), and stratified by MSKCC risk groups. Sunitinib was given at 50 mg/d, 4/6wk with dose adaptation to routine practice. In arm A, sunitinib had to start 4 to 6 wk after surgery. Primary endpoint was overall survival (OS). A total of 576 pts had to be enrolled to demonstrate non inferiority hypothesis (H0: λE/λC &gt; 1.20), with 80% power at a 1-sided significance level of 5%. Results: 450 pts were included from 9/09 to 9/17, 226 and 224 in arm A and B, respectively. Median age was 62, ECOG-PS was 0 in 56% and 1 in 44%. MSKCC risk groups were intermediate/poor in 55.6/44.4% (arm A) and in 58.5/41.5% (arm B). In arm A, 6.7% did not have CN and 22.5% never received sunitinib. In arm B, 4.9 % never received sunitinib and 17% had secondary nephrectomy. At the time of the analysis, 326 deaths have been observed with a median follow-up of 50.9 mo. OS was not inferior in arm B, overall as well as by MSKCC risk groups (table). No difference in response rate and PFS was observed. Safety of sunitinib was as expected in both arms. Conclusions: Sunitinib alone is not inferior to CN followed by sunitinib in synchronous mRCC both in intermediate and poor MSKCC risk groups. CN should not be anymore the standard of care when medical treatment is required. Clinical trial information: NCT00930033. [Table: see text]

885P - Sunitinib versus pazopanib for patients with metastatic renal cell carcinoma: Two Turkish hospital experience, a retrospective comparative case series study

Background: Pazopanib (PAZ) and Sunitinib (SUN), are two oral multikinase angiogenesis inhibitors which are prescribed frequently. However, the outcomes in real world of Turkish population have not extensively been studied. Methods: Patients assessed retrospectively at two Turkish hospitals between 2006 and 2016. Results: Median age of patients was 60 (28-87) years and 70% of patients were male. ECOG performance score was 0 and 1 in 73% of patients. Twelve patients (15%) had non-clear cell carcinoma histology. Pathological characteristics, MSKCC risk groups, median follow up, response rates and survival are shown in Table. In the SUN group, the patients had more grade 3-4 adverse events (Table).Table885P Patient characteristics, responses to treatment, survival, and adverse eventsSunitinib (n = 41)Pazopanib (n = 38)p valueMSKCC risk groupFavorable 31% Intermediate 56% Poor 12.5%Favorable 31% Intermediate 47% Poor 21%p = 0.66T3-T4 stage49%47%p = 0.38Node positivity (%)20%8%p = 0.10Local recurrence (%)20%30%p = 0.36Median follow-up18 months13 monthsp = 0.21ORR34%37%p = 0.96DCR78%87%p = 0.046Progression73% (n = 30)50% (n = 19)p = 0.08Median PFS8months8monthsp = 0.49Median OS22 months21 monthsp = 0.21Fatigue, all grades45%74%p = 0.48Skin changes, all grades44%44%p = 0.24Anemia, all grades35%42%p = 0.75Grade 3-4 adverse events59%16%p < 0.001Dose reduction50%24%p = 0.021Treatment cessation37%26%p = 0.37 Open table in a new tab Conclusions: In our study, there was no difference in terms of survival between two agents. However, patients treated with SUN had more grade 3-4 adverse effects which prompted dose reduction and cessation. Legal entity responsible for the study: Individuals, Meltem Ekenel and Senem Karabulut Funding: None Disclosure: All authors have declared no conflicts of interest.

Clinical outcomes by nephrectomy status in METEOR, a randomized phase 3 trial of cabozantinib (cabo) vs everolimus (eve) in patients (pts) with advanced renal cell carcinoma (RCC).

4570 Background: Most pts with advanced RCC undergo nephrectomy (Nx) as curative or palliative therapy. In a retrospective analysis of pts treated with targeted therapy, pts who were older and had more comorbidities and higher tumor grade were less likely to have had Nx. Pts without Nx had shorter overall survival (OS) than pts with Nx (Hanna, J Clin Oncol 2016). Here we report outcomes for cabo vs eve in pts with advanced RCC with or without prior Nx in the phase 3 METEOR trial (NCT01865747). Methods: 658 pts with clear cell RCC and ≥1 prior VEGFR TKI were randomized 1:1 to receive cabo at 60 mg qd or eve at 10 mg qd. Stratification was by MSKCC risk group and number of prior VEGFR TKIs. Endpoints included progression-free survival (PFS), OS, and objective response rate (ORR). Results: 85% of enrolled pts had prior Nx of which 7% were partial; 15% had no prior nephrectomy (NoNx). Baseline characteristics, including Karnofsky performance status (KPS), MSKCC risk group, time from diagnosis to randomization, and median sum of diameters (SoD) for tumor target lesions, were less favorable for the NoNx subgroup (Table). Improved PFS and OS with cabo vs eve were observed regardless of Nx status. For the Nx subgroup, the hazard ratio (HR) was 0.51 (95% CI 0.41-0.64) for PFS and 0.66 (95% CI 0.52-0.84) for OS; for the NoNx subgroup, the HR was 0.51 (95% CI 0.30-0.86) for PFS and 0.75 (95% CI 0.44-1.27) for OS. Median OS was longer in the Nx subgroup for both treatment arms (Table). ORR per independent radiology committee (IRC) for cabo vs eve was 17% vs 4% for Nx and 21% vs 2% for NoNx. Grade 3 or 4 adverse events for both subgroups were generally consistent with the safety profiles of cabo and eve in the overall population. Conclusions: Cabo improved PFS, ORR, and OS compared with eve in pts with advanced RCC irrespective of nephrectomy status. Clinical trial information: NCT01865747. [Table: see text]

Nivolumab treatment for patients with non-clear cell renal cell carcinoma: A multicenter retrospective analysis.

4586 Background: Nivolumab (Nivo) is approved for metastatic renal cell carcinoma (mRCC) refractory to prior antiangiogenic therapy. However, the clinical activity of Nivo in non-clear cell RCC subtypes remains unknown as these patients were not part of the trial population. Methods: Patients (pts) from 3 centers who received at least one dose of Nivo as monotherapy for mRCC between 12/2015 and 01/2017 were identified. All patients had histologically-confirmed non-clear cell histology. A retrospective analysis focusing on patient and treatment characteristics, objective response according to RECIST v1.1 and adverse events (AEs) was undertaken. Results: Twenty-three patients were identified. Median age was 59 years (33-82), 78% were male, 61% were white, 30% black and 9% Hispanic. MSKCC risk groups were 19% favorable, 67% intermediate, 14% poor. Most common histologies were Unclassified (48%) and Papillary (44%), followed by Collecting duct and Mucinous tubular and spindle cell carcinoma (MTSCC) (4% each). The majority of pts had ECOG PS 0 (57%) or 1 (33%). At diagnosis, 65% had metastatic disease; most common sites were lung (57%), retroperitoneal lymph nodes (35%), and liver or bone (26% each). Most pts (74%) had prior nephrectomy and had received one (74%) or &gt; 1 (26%) prior systemic therapy, most commonly sunitinib (65%), pazopanib (30%) or axitinib (17%). No pts had prior IL-2, but one pt had atezolizumab on trial. After a median follow-up of 6.5 months (mos) and 8 Nivo doses, median PFS was 4.2 mos and median OS was not reached. Among 21 pts evaluable for best response, 6 (29%) had PR and 4 (19%) had SD. Median time to best response was 5.1 mos. At the time of analysis, 9 pts were still on Nivo, 2 were on treatment break, 8 pts had subsequent treatment after PD on Nivo, and 2 were lost to follow-up. Two of 23 pts died. Most common AEs ( &gt; 10%, any grade) were fever (13%) and fatigue (13%). Nivo treatments were postponed in 6 pts due to intolerance and stopped in3 pts. There were no treatment-related deaths. Conclusions: Nivolumab monotherapy demonstrated a substantial objective response rate and was well tolerated in a heterogeneous population of treatment-refractory patients who had mRCC with non-clear cell histology.

Prognostic value and optimal cut off of neutrophil/lymphocyte ratio (NLR) in patients with metastatic renal cell cancer (mRCC).

e16059 Background: Several biomarkers were tested in mRCC but only outcome scores showed prognostic significance. Neutrophil-to-lymphocyte ratio (NLR) has shown prognostic significance for OS in a number of cancer types but value and cutoff in mRCC is unknown. Methods: The aim of this study was to evaluate NLR as prognostic factor and find a cutoff in mRCC pts. The medical records of 103 pts with mRCC were reviewed. Analysis included adjusting for baseline disease and patient characteristics and investigating potential interaction effects between NLR status and significant BL predictors of outcome. Overal Survivall (OS) was estimated by the Kaplan-Meier method.Univariate and multivariate analysis was performed by Cox regression. Results: Total pts reviewed were 103, MSKCC risk groups (37% Good, 47% Intermediate,16% Poor).Median follow up were 50 months and median OS 29 months. Nephrectomy was performed in 81% of cases.Ninety six percent received treatment (tki – mTor inhibitors). The optimal cutoff was 4, pts with NLR ≤4 vs ≥4 have higher OS 32m vs 10m HR 3,38 (IC95%: 2,60- 4,50 p:0,0001) univariate and HR: 4,32 (IC 95%: 3,35-5,4 p:0,003) in the multivariate analysis. Conclusions: Our series show MSKCC score and RLN were prognostic factors in univariable and multivariable analisys. NRL ≥ 4 identify a group of patients with poor overall survival. The NLR provides independent prognostic information for patients with mRCC and could be useful in academic and community setting. [Table: see text]

Outcomes based on age in the phase 3 METEOR trial of cabozantinib (cabo) vs everolimus (eve) in patients with advanced renal cell carcinoma (RCC).

4578 Background: The incidence of RCC increases with age with the highest incidence at ~75 years of age (Znaor, Eur Urol 2015). The Phase 3 METEOR trial (NCT01865747) showed a significant improvement in progression-free survival (PFS; HR 0.58, 95% CI 0.45–0.74; P &lt; 0.0001), overall survival (OS; HR 0.66, 95% CI 0.53–0.83, P = 0.0003), and objective response rate (ORR; 17% vs 3%; P &lt; 0.0001) for cabo compared with eve in patients with advanced RCC previously treated with VEGFR TKIs (Choueiri, NEJM 2015, Lancet Oncol 2016). Here we present outcomes by 3 categories of age for the METEOR trial. Methods: 658 patients were randomized 1:1 to cabo (60 mg qd) or eve (10 mg qd). Stratification factors were MSKCC risk group and number of prior VEGFR TKIs. Endpoints included PFS, OS, and ORR. Subgroup analyses by age ( &lt; 65, 65 to 74, and ≥75 years) are presented. Results: At baseline, 60% of patients were &lt; 65 years old, 31% were 65 to 74 years old, and 10% were ≥75 years old. Subgroups by age generally had similar baseline characteristics in both arms. The HRs for PFS favored cabo for all age groups (HR 0.53, 95% CI 0.41–0.68 for &lt; 65 years old; 0.53, 95% CI 0.37–0.77 for 65 to 74 years old; and 0.38, 95% CI 0.18–0.79 for ≥75 years old). ORR per independent radiology committee for cabo vs eve was 15% vs 5% for &lt; 65 years old, 21% vs 2% for 65 to 74 years old, and 19% vs 0% for ≥75 years old. HRs for OS also favored cabo (HR 0.72, 95% CI 0.54–0.95 for &lt; 65 years old; 0.66, 95% CI 0.44–0.99 for 65 to 74 years old; and 0.57, 95% CI 0.28–1.14 for ≥75 years old). Median OS for cabo vs eve was 21.4 mo vs 17.1 mo for &lt; 65 years old, not reached vs 18.0 mo for 65 to 74 years old, and 18.4 mo vs 14.0 mo for ≥75 years old. Older patients more frequently had dose reductions (60% with cabo and 22% with eve for &lt; 65 years old vs 85% with cabo and 36% with eve for ≥75 years old). Grade 3 or 4 adverse events were generally consistent with the safety profiles in the overall population although some events such as fatigue and hypertension occurred at a higher rate in older patients. Conclusions: Treatment with cabo improved PFS, ORR, and OS compared with eve in patients with advanced RCC irrespective of age. Adverse events in older patients were more frequently managed with dose reductions. Clinical trial information: NCT01865747.