Clinical outcomes and ctDNA correlates for CAPOX BETR: a phase II trial of capecitabine, oxaliplatin, bevacizumab, trastuzumab in previously untreated advanced HER2+ gastroesophageal adenocarcinoma

Abstract

Preclinical studies suggest that simultaneous HER2/VEGF blockade may have cooperative effects in gastroesophageal adenocarcinomas. In a single-arm investigator initiated clinical trial for patients with untreated advanced HER2+ gastroesophageal adenocarcinoma, bevacizumab was added to standard of care capecitabine, oxaliplatin, and trastuzumab in 36 patients (NCT01191697). Primary endpoint was objective response rate and secondary endpoints included safety, duration of response, progression free survival, and overall survival. The study met its primary endpoint with an objective response rate of 81% (95% CI 65-92%). Median progression free and overall survival were 14.0 (95% CI, 11.3-36.4) and 23.2 months (95% CI, 16.6-36.4), respectively. The median duration of response was 14.9 months. The regimen was well tolerated without unexpected or severe toxicities. In post-hoc ctDNA analysis, baseline ctDNA features were prognostic: Higher tumor fraction and alternative MAPK drivers portended worse outcomes. ctDNA at resistance identified oncogenic mutations and these were detectable 2-8 cycles prior to radiographic progression. Capecitabine, oxaliplatin, trastuzumab and bevacizumab shows robust clinical activity in HER2+ gastroesophageal adenocarcinoma. Combination of VEGF inhibitors with chemoimmunotherapy and anti-PD1 regimens is warranted.

Fig. 1. Consort diagram.

Patient flowchart showing selection of cohort included in the trial.

Fig. 2. Clinical outcomes with CAPOX, trastuzumab and bevacizumab in patients with previously untreated advanced HER2 + GEA.

A Waterfall Plot showing Best Response per RECIST 1.1. B Spider plot demonstrates evolution of radiographic response over time as measured by the change in sum of target lesions measured by RECIST 1.1 (N = 36). Progression free survival (PFS, C) and Overall survival (OS, D), respectively, for the entire cohort. Progression free survival E and Overall survival F, respectively, stratified by subjects with tumors that were IHC HER2 3+ vs. IHC HER2 2+ (N = 33). P-value was calculated using two-tailed log rank test E, F.

Fig. 3. Fluorescent in situ hybridization (FISH) outlines spectrum of HER2 heterogeneity.

A Fraction of cells displaying HER2 amplifications as determined by single cell HER2 FISH shows significant heterogeneity. A minimum of 50 tumor nuclei were measured to determine HER2 amplified fraction (y-axis) (N = 14) B Representative images from HER2 FISH showing absent (left), intermediate (middle) and high levels (right) of HER2 heterogeneity. Red probe binds HER2 locus, and green probe binds centromeric region of chromosome 17 (CEP17). Ratio of HER2/CEP17 is used to identify HER2 amplification and polysomy chromosome 17. Middle panel: Normal tissue provides contrast with no evidence of HER2 amplification. Blue dashed circles: HER2 negative cells, Red dashed circle: HER2 positive cells. C. HER2 2 + IHC (n = 3) tumors show significantly higher HER2 heterogeneity or lower HER2 amplified fraction compared to HER2 3 + IHC (n = 10). Dots represent individual patients, box extends from 25th to 75th percentile, horizontal line within box represents median, and whiskers extend a maximum of 1.5 x interquartile range (IQR). A, C. Groups compared using two-tailed Wilcoxon rank-sum test C.

Fig. 4. Baseline ctDNA genomic analysis identifies prognostic biomarkers.

A A CLIA-approved 152-gene next generation sequencing (NGS) assay (Predicine CARE^TM) was used to perform ctDNA based genomic profiling on diagnostic &/or pre-treatment plasma samples from 30 cases in our clinical trial. Genomic profiling showed high rates of TP53 alterations and HER2 amplifications were identified in 76.7% of the cohort (23/30). B 56.7% of the cases in our cohort (17/30) have evidence of additional MAPK driver alteration in addition to HER2. Amplifications in the receptor tyrosine kinases (RTKs) including in EGFR, KRAS, MET, and FGFR1 are the most common additional MAPK alterations. Cases with additional MAPK alterations (MAPK Alt) have a significantly worse progression free C and overall survival D compared to those who’s tumors without additional MAPK alterations (MAPK Wild Type, WT). Patients with high tumor fraction (TF) at diagnosis, estimated using a combination of low pass whole genome sequencing (LP-WGS) using ichorCNA and somatic mutation allele frequencies in ctDNA NGS assay, display a significantly worse progression free E and overall survival F. P-value was calculated using two-tailed log rank test C–F. BRR Best radiographic response, CR Complete Response, PR Partial Response, SD Stable disease, PD Progressive disease.

Fig. 5. Early ctDNA changes after one week of trastuzumab monotherapy predict outcomes.

A Initial therapeutic strategy on this trial included trastuzumab monotherapy which was followed by the complete regimen being given 7 days later. Tumor fraction at time of trastuzumab monotherapy and one week post are shown. Cases with % ctDNA decrease > median are in blue, and those with % ctDNA decrease <median or ctDNA increase are in red (n = 11 each). Cases with tumor fraction <0.05 (horizontal dashed line) were excluded from the analysis. B Cases with higher drop in ctDNA (blue) after one week of trastuzumab monotherapy have trend towards improved PFS. P-value was calculated using two-tailed log rank test B.

Fig. 6. Prevalence and dynamics of putative resistance alterations in longitudinal ctDNA analysis.

A Comparison of ctDNA NGS profiles from diagnosis and at the time of progression on therapy in 19 cases are shown. 10 cases show evidence of novel alterations (orange boxes) including in several putative resistance alleles such as KRAS, NRAS, PIK3CA, and PTEN. Interestingly, Pt 9 shows evidence of multiple new ERBB2 alterations. B Longitudinal monitoring of dynamics of resistance (red) and truncal alleles (black) in select four cases using ultra-deep sequencing for higher sensitivity. In general, low level resistance alleles are identified at diagnosis, are effective suppressed with therapy re-emerging several cycles prior to radiographic progression.

Fig. 7. Tumor agnostic ctDNA methylation profiling reveals concordant tumor fraction changes to mutation based ctDNA approach.

Comparison of ctDNA derived tumor burden over course of therapy in index cases ID9 A and ID15 B using average mutation allele frequency (MAF) derived from ctDNA based MRD assay (black), and ctDNA based tumor agonistic methylation panel using differential methylation fragment burden (DMFs, red) shows similar dynamics and confirms molecular ctDNA progression precedes radiographic progression. C Heatmap of changes in signal over 2396 cancer associated methylation regions over the course of therapy with re-emergence of signal starting cycle 16 of chemotherapy in ID15. D Fifteen loci known to be hypermethylated in gastric cancer showing high methylation signature in ctDNA at baseline and changes of response and re-emergence around cycle 16 in ID15.