Ramucirumab, Avelumab, and Paclitaxel as Second-Line Treatment in Esophagogastric Adenocarcinoma: The Phase 2 RAP (AIO-STO-0218) Nonrandomized Controlled Trial

Importance

Adding immune checkpoint inhibitors to chemotherapy has been associated with improved outcomes in metastatic esophagogastric adenocarcinoma, but treatment combinations and optimal patient selection need to be established.

Objective

To investigate the efficacy and tolerability of the programmed cell death ligand 1 (PDL-1) inhibitor avelumab with paclitaxel plus ramucirumab.

Design, Setting, and Participants

This multicenter, single-group, phase 2 nonrandomized controlled trial was conducted among patients with second-line metastatic esophagogastric adenocarcinoma. Patients pretreated with platinum plus fluoropyrimidine between April 2019 and November 2020 across 10 German centers (median follow-up, 27.4 months [95% CI 22.0-32.9 months]) were included. Data analysis was performed from January to December 2022.

Interventions

Patients received ramucirumab at 8 mg/kg on days 1 and 15, avelumab at 10 mg/kg on days 1 and 15, and paclitaxel at 80 mg/m2 on days 1, 8, and 15 every 4 weeks.

Main Outcomes and Measures

The prespecified primary end point was overall survival (OS) rate at 6 months, with the experimental therapy considered insufficiently active with an OS rate of 50% or less and a promising candidate with an OS rate of 65% or greater.

Results

Of 60 enrolled patients, 59 patients (median [range] age, 64 [18-81] years; 47 males [70.7%]) were evaluable, including 30 patients with metastatic adenocarcinoma of the stomach and 29 patients with gastroesophageal junction. All patients were pretreated with platinum plus fluoropyrimidine, and 40 patients (67.8%) had received prior taxanes; 24 of 56 evaluable patients (42.9%) had a PDL-1 combined positive score (CPS) of 5 or greater, centrally assessed. The OS rate at 6 months was 71.2% (95% CI, 61.5%-83.7%). The median OS in the intention-to-treat population (59 patients) was 10.6 months (95% CI, 8.4-12.8 months) overall. Among patients assessable by central pathology, median OS was 9.4 months (95% CI, 7.2-11.7 months) in 32 patients with a PDL-1 CPS less than 5 and 14.0 months (95% CI, 6.0-22.1 months) in 24 patients with a PDL-1 CPS of 5 or greater (P = .25). Treatment was generally well tolerated, without unexpected toxicities. Patients with higher vs lower than median T cell repertoire richness showed an increased median OS of 20.4 months (95% CI, 7.7-33.0 months) compared with 8.3 months (95% CI, 3.7-12.9 months; hazard ratio, 0.43; 95% CI, 0.23-0.81; P = .008). Patients with lower vs higher than median cell-free DNA burden had a median OS of 19.2 months (95% CI, 8.9-29.6 months) compared with 7.3 months (95% CI, 3.2-11.4 months; hazard ratio, 0.30; 95% CI, 0.16-0.59; P < .001).

Conclusions and relevance

In this study, the combination of avelumab with paclitaxel plus ramucirumab showed favorable efficacy and tolerability in the second-line treatment for metastatic esophagogastric adenocarcinoma. A PDL-1 CPS score of 5 or greater, cell-free DNA level less than the median, and T cell repertoire richness greater than the median were associated with increased median OS.

Trial Registration

ClinicalTrials.gov Identifier: NCT03966118.

Recent progress and current challenges of immunotherapy in advanced/metastatic esophagogastric adenocarcinoma

The new era of immunotherapy is successfully implemented in the treatment of metastatic/locally advanced esophagogastric adenocarcinoma (EGAC), as it has been investigated in combinations with/without chemotherapy in human epidermal growth factor receptor 2 (Her2)-positive and Her2-negative tumors. Recent approvals of immune checkpoint inhibitors (ICI) enrich the therapeutic landscape in nearly every therapeutic line. Based on CHECKMATE-649, the combination of nivolumab and chemotherapy in first-line therapy of programmed cell death protein 1 (PD-L1)-positive patients with advanced gastroesophageal junction cancer (GEJC), esophageal cancer (EC), and gastric cancer (GC) was approved in Europe for PD-L1 combined positivity score (CPS) ≥ 5 patients and independently from PD-L1 score in the USA and Asia. Based on KEYNOTE-590, patients with advanced GEJC and EC qualify for the combination of pembrolizumab plus chemotherapy in Europe (CPS ≥ 10) and the USA. For Her2-positive patients, trastuzumab with first-line chemotherapy plus pembrolizumab has beneficial response rates and resulted in approval in the USA (KEYNOTE-811). In third-line therapy, superior overall survival (OS) was achieved by the administration of nivolumab (approval in Japan, ATTRACTION-02), and pembrolizumab shows a positive effect on the duration of response (KEYNOTE-059). Questions of resistance to immunotherapy or the role of gender in response to ICI need to be clarified. This review provides an overview of the current approvals of ICI in advanced EGAC and reflects results of relevant phase II/III trials with focus on possible biomarkers, including PD-L1 CPS and microsatellite-instability (MSI) status.

Ramucirumab, avelumab, and paclitaxel (RAP) as second-line treatment in gastro-esophageal adenocarcinoma, a phase II trial of the Arbeitsgemeinschaft Internistische Onkologie (AIO)

4051

Background: Combination of ramucirumab and paclitaxel is a standard second line therapy in gastro-esophageal adenocarcinoma (GEAC) (RAINBOW trial, Wilke et al., 2014). We integrated the PD-L1 inhibitor avelumab into this regimen aiming for synergistic efficacy. Methods: In a multicenter phase II trial (NCT03966118) pts with metastatic GEAC, after progression on platinum / fluoropyrimidine based palliative 1st-line, ECOG 0 or 1, were treated with ramucirumab 8 mg/kg (d1,15) + avelumab 10 mg/kg (d1,15) + paclitaxel 80 mg/m² (d1,8,15), q4w. Sample size calculation was based on a Simon 2-stage design with overall survival rate at 6 months as the primary endpoint (H0≤50%, H1≥65%). Results: 60 pts were enrolled, 59 were evaluable (ITT), median age 64.0 yrs (range 18-81), male 80%, female 20%, primary gastric 52%, GEJ 48%, histology intestinal 59.6%, diffuse 24.6%, mixed 15.8%. Previous treatment with platin/fluoropyrimidine 100%, previous taxanes 66%. At central pathology MSI-H 7%, PD-L1 CPS < 5: 54%; ≥5: 41%, NA 5%. Response by investigator (%) CR 3.4, PR 27.1, SD 49.2, PD 20.3; DCR 79.7%, independent radiology review will be presented. DOR: 8.2 mo (95%CI 6.7-9.7), PFS 5.4 mo (95%CI 4.2-6.6); 6-mo OS rate 71.2%; 12-mo OS rate 45.8%; med OS (ITT) 10.6 mo (95%CI 8.2-13.1), med OS CPS < 5: 9.4 mo (95%CI 7.2-11.2), CPS ≥5: 14.0 mo (95%CI 12.8-15.3), translational data (ct-DNA) will be analysed. Treatment was generally well tolerated and no unexpected toxicities occurred: Grade 3/4 AE above 5%: anemia 5%, leucopenia 12%, neutropenia 22%, diarrhea 5%, pain 10%, peripheral neuropathy 10%, hypertension 5%, non-neutropenic infection 5% including 1 CTC grade 5 due to an esophago-tracheal fistula. Conclusions: The med OS of 10.6 mo (in a population of 66% pretreated with taxanes) compares very favourably to 8.6 mo in the Western population RAINBOW trial (Shitara et al., 2016) and 7.6 mo in the RAMIRIS trial (Lorenzen et al., 2022). PD-L1 CPS≥ 5 seems to predict for an even better efficacy (med OS 14.0 mo). Second-line RAP is a very efficacious and well tolerated combination. Clinical trial information: NCT03966118.

Immunotherapy in Gastric Cancer

Immune checkpoint inhibition is a new standard of targeted therapy in the treatment of advanced or metastatic gastric cancer (GC) and is represented in various combinations with and without chemotherapy in every therapy line within clinical trials. In advanced adenocarcinoma of GC, gastroesophageal junction cancer (GEJC) and esophageal cancer (EC), the combination of nivolumab and chemotherapy in first-line therapy improves overall survival (OS) in PD-L1 (programmed cell death protein 1)-positive patients with approval in Europe (PD-L1 CPS (combined positivity score) ≥ 5), USA and Taiwan (CHECKMATE-649) and pembrolizumab plus chemotherapy for GEJC and EC in Europe (CPS ≥ 10) and the USA (KEYNOTE-590). Furthermore, pembrolizumab plus trastuzumab and chemotherapy show clear benefits in OS and are approved as first-line treatment of Her2 (human epidermal growth factor receptor-2)-positive tumors in the USA (KEYNOTE-811). Nivolumab demonstrates superior OS regardless of PD-L1 expression in third-line therapy with approval in Japan (ATTRACTION-02) and pembrolizumab prolonged the duration of response in PD-L1 positive patients with approval in the USA in PD-L1 CPS ≥ 1 patients (KEYNOTE-059). This review reflects the rationale and current results of phase II and III clinical trials investigating various immune checkpoint inhibitors targeting PD-L1/1 and CTLA (anticytotoxic T-lymphocyte-associated antigen)-4 in combination with and without chemotherapy and Her2-targeted therapy in GC.

Pazopanib with 5-FU and oxaliplatin as first line therapy in advanced gastric cancer: A randomized phase-II study-The PaFLO trial. A study of the Arbeitsgemeinschaft Internistische Onkologie AIO-STO-0510

VEGF inhibition in gastric cancer has a proven benefit in the second line setting. Pazopanib, an oral tyrosine kinase inhibitor, selectively inhibits VEGFR-1, -2 and -3, c-kit and PDGF-R resulting in inhibition of angiogenesis. This open-label randomized phase II trial (2:1) investigated the efficacy of combining pazopanib with FLO (5-fluorouracil, oxaliplatin) vs FLO alone (internal control arm) as first-line treatment in patients with advanced adenocarcinoma of the stomach and gastroesophageal junction (GEJ). Eighty-seven patients were randomized and 78 patients were eligible and evaluable (PaFLO arm 51 patients, FLO arm 27 patients). The PFS rate at 6 months (primary endpoint) was 34% in the PaFLO arm vs 30% in the FLO arm. Comparing PaFLO with FLO median PFS was 4.66 months (95% confidence interval [CI] 2.87-6.46) vs 4.47 months (95% CI 1.79-7.14) (95% CI, hazard ratio [HR] 0.96 (0.60-1.55), P = .882 [exploratory]); median OS was 10.19 months (95% CI 5.46-14.92) vs 7.33 months (95% CI 4.93-9.73), (95% CI HR 1.01 [0.62-1.65], P = .953, exploratory), disease control rate was 72% vs 59%. PaFLO was well tolerable, toxicities were slightly higher in the PaFLO arm. Major adverse events were loss of appetite, nausea, fatigue, diarrhea, neutropenia and thrombocytopenia. Adding pazopanib to chemotherapy shows signs of efficacy but no major improvement in this randomized phase 2 trial. The PFS at 6 months in both arms was lower than expected from the literature. Biomarkers identifying subgroups who benefit and novel combinations are needed. ClinicalTrials.gov: NCT01503372.

Immune Checkpoint Inhibition in Oesophago-Gastric Carcinoma

Immune checkpoint inhibitors enrich the therapeutic landscape in oesophago-gastric carcinoma. With regard to oesophageal squamous cell carcinoma (ESCC), the selective PD-1 (programmed cell death receptor 1)-inhibitor nivolumab improves disease-free survival in the adjuvant therapy setting (CHECKMATE-577). In first-line treatment, ESCC patients (pts) benefit in overall survival (OS) from the PD-1-inhibitor pembrolizumab in combination with chemotherapy (KEYNOTE-590). In the second-line setting, nivolumab (ATTRACTION-03) and pembrolizumab (KEYNOTE-181) demonstrate a benefit in OS compared with chemotherapy. These data resulted in the approval of nivolumab for the second-line treatment of advanced ESCC pts regardless of PD-L1 (programmed cell death ligand 1) status in Europe, Asia, and the USA, and pembrolizumab for pts with PD-L1 CPS (combined positivity score) ≥ 10 in Asia and the USA. Further approvals can be expected. In gastro-oesophageal junction and gastric cancer, the addition of nivolumab to chemotherapy in first-line treatment improves OS in pts with advanced disease with PD-L1 CPS ≥ 5 (CHECKMATE-649). Additionally, pembrolizumab was non-inferior to chemotherapy for OS in PD-L1 CPS ≥ 1 pts (KEYNOTE-062). In third-line treatment, nivolumab shows benefits in OS regardless of PD-L1 expression (ATTRACTION-02) with approval in Asia, and pembrolizumab prolonged the duration of response in PD-L1 positive pts (KEYNOTE-059) with approval in the USA. We discuss the recent results of the completed phase II and III clinical trials.

Feasibility and effects of home parenteral nutrition (HPN) in cancer patients with or without migrant background

e24137

Background: Parenteral nutrition is used in cancer patients (pts) requiring intensive nutrition support due to insufficient dietary intake and consuming tumor disease. As there is no data of HPN in pts with migrant background (MB), this observational study examined viability of HPN, influence on QoL and nutrition status of cancer pts with and without MB. Methods: Eligibility included ECOG ≥ 1, life expectancy ≥ 4 weeks, first time HPN for at least 28 days and written informed consent. Migrant background status and patient characteristics were assessed at baseline. At day 1 (baseline), day 14 and day 28 validated questionnaires assessed QoL, nutrition status (EORTC-QLQ-C15-PAL, Subjective global assessment (SGA)), as well as feasibility and complications of HPN. Additionally, bioimpedance analysis, BMI and weight were measured to evaluate response of HPN. Results: Between Mai 2015 and November 2019, 68 pts were included, 17 of them with MB. Tumor entities were gastric (n = 41), esophageal (n = 20) and other (n = 7), treated in a curative (n = 14) or palliative (n = 54) concept. Overall, 40 pts were able to continue HPN until day 14, 26 pts until day 28. 42 pts left study due to death (1 with MB/8 overall), worsening of general condition (3/15), dietary improvement (1/2), change of supplier (1/5), informed consent withdrawal (0/1) and lost to follow up (2/11). Global QoL in all pts was stable from baseline to d14 (n = 36; 36.6 to 40.3), whereas QoL in pts with MB deteriorated (n = 9; 37.0 to 27.8). From baseline to d28, QoL in all pts improved (n = 23; 38.4 to 47.8) and in pts with MB QoL was stable (n = 7; 40.5 to 42.9). Medical problems associated with HPN occurred in 4/11 pts with MB (36%) and 11/37 pts without MB (30%), respectively. The majority of pts reported about HPN-associated organizational complications, whereas there is no significant difference of pts with MB (6/10) and without MB (20/38). SGA improvement was observed in all pts regardless of MB. During the course of HPN, there was no significant change in body weight and BMI in all pts, independent from MB. Conclusions: Overall, 59% of pts received HPN for 14 days, only 38% of pts for 28 days. Discontinuation was mainly due to disease progression. Feasibility and organizational problems did not affect duration of HPN. Assessing HPN prospectively shows, that defining the ideal point to start HPN appears to be essential for response to HPN. In our study, migrant background is not associated with a major difference in incidence of complications, viability, benefit or duration of HPN.

RAP: A phase II trial with ramucirumab, avelumab, and paclitaxel as second line treatment in gastro-esophageal adenocarcinoma of the arbeitsgemeinschaft internistische onkologie (AIO)

TPS4148

Background: Combination of ramucirumab and paclitaxel resembles the standard treatment option in second line therapy with improvement of response rate and overall survival (REGARD, RAINBOW). Response rates to PD-1/L1 blockade in gastro-esophageal cancer patients rank within 10–20%, whereby PD-1/L1 blockade is reported to impressively extend survival rates in responders. Trials investigating either the synergistic effect of anti-angiogenesis and anti-PD-L1 or chemotherapy combined with anti-PD-L1 are promising. Based on these data we hypothesize benefit from combining immunotherapy by checkpoint inhibition with VEGF-directed treatment and chemotherapy induced increase of immunogenicity of tumor cells. This study investigates the incorporation of PD-L1 blockade by avelumab in the second line setting by combination with the actual best second-line chemotherapy regimen in metastatic gastric cancer patients (paclitaxel+ramucirumab). Methods: The RAP trial (AIO-STO-0218, registered at ClinicalTrials.gov) is a single arm multicenter phase II trial. A total of 59 patients with metastatic or locally advanced gastric or gastro-esophageal junction adenocarcinoma, ECOG 0–1, who progressed after having received first-line therapy with platinum and fluoropyrimidine doublet with or without anthracycline, docetaxel or trastuzumab within the last six months will receive avelumab and ramucirumab on day 1, 15 and paclitaxel on day 1, 8 and 15 of a 28-day cycle until disease progression (RECIST v1.1), intolerable toxicity, withdrawal of consent or at a maximum treatment of 1 year. The primary endpoint is the overall survival rate (OSR) at 6 months. Sample size calculation is based on a Simon 2-stage design with a one-sided alpha error of 10% and a power of 80%, an expected OSR at 6 months of ≥ 65% and a 0 hypothesis ≤ 50%. Secondary endpoints include OS, OSR at 12 months, PFS, safety and tolerability, duration of response. Ethics commission approved the study protocol in January 2019. Updated patient accrual will be presented. Clinical trial information: AIO-STO-0218.

Myoglobin facilitates angiotensin II-induced constriction of renal afferent arterioles

Vasoconstriction plays an important role in the development of acute kidney injury in rhabdomyolysis. We hypothesized that myoglobin enhances the angiotensin II (ANG II) response in afferent arterioles by increasing superoxide and reducing nitric oxide (NO) bioavailability. Afferent arterioles of C57Bl6 mice were isolated perfused, and vasoreactivity was analyzed using video microscopy. NO bioavailability, superoxide concentration in the vessel wall, and changes in cytosolic calcium were measured using fluorescence techniques. Myoglobin treatment (10−5 M) did not change the basal arteriolar diameter during a 20-min period compared with control conditions. NG-nitro-l-arginine methyl ester (l-NAME, 10−4 M) and l-NAME + myoglobin reduced diameters to 94.7 and 97.9% of the initial diameter, respectively. Myoglobin or l-NAME enhanced the ANG II-induced constriction of arterioles compared with control (36.6 and 34.2%, respectively, vs. 65.9%). Norepinephrine responses were not influenced by myoglobin. Combined application of myoglobin and l-NAME further facilitated the ANG II response (7.0%). Myoglobin or l-NAME decreased the NO-related fluorescence in arterioles similarly. Myoglobin enhanced the superoxide-related fluorescence, and tempol prevented this enhancement. Tempol also partly prevented the myoglobin effect on the ANG II response. Myoglobin increased the fura 2 fluorescence ratio (cytosolic calcium) during ANG II application (10−12 to 10−6 M). The results suggest that the enhanced afferent arteriolar reactivity to ANG II is mainly due to a myoglobin-induced increase in superoxide and associated reduction in the NO bioavailability. Signaling pathways for the augmented ANG II response include enhanced cytosolic calcium transients. In conclusion, myoglobin may contribute to the afferent arteriolar vasoconstriction in this rhabdomyolysis model.

Abstract 3181: Expression changes of the chromatin modifier PBRM1 in human renal cell carcinomas in relation to histopathological features

Recent next-generation sequencing studies of clear cell renal cell carcinoma (ccRCC) have identified point mutations in chromatin-modifying genes, among them PBRM1 which is now the second most frequently mutated gene in RCCs (in up to 41% ccRCCs cases) and belongs to the m1 TCGA (The Cancer Genome Atlas) subset. The occasional presence of PBRM1 mutations in the absence of VHL mutations indicates, that PBRM1 might be another driver tumor suppressor gene in ccRCCs. We were interested in examining PBRM1 expression at mRNA and protein levels and correlated these findings to histopathological features of RCC patients.

Relative gene expression (RGE) data were obtained by Q-PCR from matched tumor and adjacent normal fresh frozen ccRCC tissues from 57 patients who underwent radical nephrectomy at our clinic. Papillary (pRCC) as well as chromophobic cases (chRCC) were added to analyze the expression of splice variants of PBRM1. Comparative Western blot analyses as well as semi-quantitative immuno-histochemical staining (IHC) for PBRM1 and VHL were performed on all ccRCC cases.

In 78.9% of cases (45/57) PBRM1 mRNA was downregulated at least 1.5-fold, 7% (4/57) showed PBRM1 upregulation, and 14% (8/57) displayed no obvious expression changes between tumor and corresponding tumor tissue. Interestingly, 21 of 45 ccRCCs tumor tissues expressed splice variant 4 more abundantly when compared to normal tissue, whereas the normal tissue preferentially expressed splice variant 1 that includes an additional exon. Remarkably, this differential expression picture is completely reversed in pRCC and chRCC cases.

The majority of 57 ccRCCs displayed weak nuclear PBRM1 staining (52.6%), whereas 31.6% showed moderate and 15.8% strong staining. However, we were not able to demonstrate a significant correlation of IHC expression levels, neither to tumor staging nor to Fuhrman grading.

The observed high frequency of decreased expression of the chromatin-remodeling gene PBRM1 on mRNA (78.9%) and protein levels (52.6%), respectively, as wells as a high mutation rate (about 15% in our data set) indicate a substantial role of PBRM1 in the tumorigenesis of ccRCCs. Preferences in the expression of different PBRM1 splice variants warrant further investigation with regard to renal cell carcinoma development.

Citation Format: Hans Krause, Anica Högner, Burkhard Jandrig, Ergin Kilic, Carsten Kempkensteffen. Expression changes of the chromatin modifier PBRM1 in human renal cell carcinomas in relation to histopathological features. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3181.