ASCO 2023 Meeting Report — Gastrointestinal Cancers

The 2023 American Society of Clinical Oncology (ASCO) annual meeting, held June 2 to 6 in Chicago, highlighted key advances in treatment across a broad spectrum of malignancies. Here, Associate Editor Dr. David Ilson reviews important trials in colorectal, gastric, and biliary tract cancers, some practice changing.

At the Plenary Session, researchers presented results from the randomized, phase 3 PROSPECT trial comparing neoadjuvant FOLFOX against chemoradiotherapy in patients with clinical stage T2N1 or T3N0–1 locally advanced mid- or high-rectal cancer who were candidates for sphincter-sparing surgery (abstract LBA2 and N Engl J Med 2023 Jun 4; [e-pub]). Patients received either standard preoperative chemoradiotherapy (5FU or capecitabine), surgery, and adjuvant chemotherapy or induction chemotherapy with six cycles of FOLFOX followed by assessment of tumor response: Patients with 20% or greater tumor reduction could proceed to surgery — without radiotherapy — followed by adjuvant chemotherapy, whereas patients with nonresponding tumors received chemoradiotherapy followed by surgery and adjuvant chemotherapy.

Of 1128 patients, median age was 57, most were male, 91% had T3N0–1 cancers, 53% had T3N1 cancers, and 64% had cancers 5 to 10 cm from the anal verge. The primary endpoint of noninferior disease-free survival at 5 years was met for induction FOLFOX compared with chemoradiotherapy (80.8% and 78.6%; hazard ratio, 0.92; P=0.005). The FOLFOX and chemoradiotherapy groups also had similar overall survival (89.5% and 90.2%; HR, 1.04) and rates of local recurrence (1.6% and 1.8%; HR, 1.18). Only 9.1% of patients in the FOLFOX arm required chemoradiotherapy, and an additional 1.4% received postoperative chemoradiotherapy. Rates of pathologic complete responses to FOLFOX and chemoradiotherapy were also similar (21.9% and 24.3%).

Comment: The PROSPECT trial indicates that 90% of patients with clinical stage T2N1 or T3N0–1 mid- or high-rectal cancers can avoid radiation therapy if given induction therapy with FOLFOX prior to surgery. This trial is practice changing. Increasingly, use of preoperative radiotherapy for rectal cancer will largely be limited to low-rectal cancers as part of total neoadjuvant therapy, particularly in patients who may be candidates for nonoperative management.

In the phase 3 NEOCOL trial, patients in Scandinavia with clinical stage T3–4 colon cancer as assessed by computed tomography imaging were randomized to either standard surgery followed by adjuvant chemotherapy as dictated by surgical pathology or preoperative chemotherapy (3 cycles of capecitabine/oxaliplatin or 4 cycles of mFOLFOX6) followed by surgery, followed by adjuvant chemotherapy as dictated by surgical pathology (abstract LBA3503).

The 250 patients had a median age of 66 years, 54% had left-sided primaries, and 73% had clinical stage T3 disease. The preoperative-therapy and up-front surgery groups did not differ in the percentage of patients proceeding to surgery (98% and 99%) and achieving a negative margin R0 resection (90% and 93%), and there was no significant difference in surgical complications. There was also no difference between groups in the primary endpoint of disease-free survival (P=0.94) or in overall survival (P=0.95). In the up-front surgery group, 27% of patients were deemed not to require adjuvant therapy.

Comment: This trial adds to the controversy about the role of preoperative therapy in colon cancer, with the results suggesting that a significant percentage of patients are clinically over-staged and may not require adjuvant chemotherapy at all.

In HER2-positive metastatic colorectal cancer, the industry-funded, randomized, phase 2 DESTINY-CRC02 trial evaluated the HER2 antibody–drug conjugate trastuzumab deruxtecan at two doses: 5.4 mg/kg or 6.4 mg/kg every 3 weeks (abstract 3501). Of the 122 patients, 80% had immunohistochemistry (IHC) 3+ tumors, 16% harbored RAS mutations, and 22% had received prior HER2-directed therapy.

In total, 82 patients received 5.4 mg/kg and 40 received 6.4 mg/kg trastuzumab deruxtecan. The objective response rate — the primary outcome — was numerically, but not statistically, higher with the lower dose compared with the higher dose (37.8% vs. 27.5%). The median duration of response was similar in the two arms (5.5 months), as was the median progression-free survival (5.5 and 5.8 months, respectively). Median overall survival was 13.3 months with the lower dose and not reached with the higher dose. In the lower-dose cohort, responses were seen in both RAS wild-type and RAS mutant cancers and in patients with prior anti-HER2 treatment, and response rates were higher in patients with IHC 3+ compared with IHC 2+/ISH+ cancers (46.9% vs. 5.6%). Rates of grade 3/4 toxicity were lower with the lower dose. There was one fatal case (2.6%) of interstitial lung disease with the higher dose.

Comment: These striking results support use of the lower dose of trastuzumab deruxtecan and further support the use of HER2-targeted agents in HER2+ colon cancer. Unique to trastuzumab deruxtecan is the retention of activity in RAS-mutant cancers, as studies of other agents found that RAS mutation conveyed resistance to HER2-targeted therapy.

In the industry-funded, phase 3 ATTRACTION 5 trial, researchers in Asia evaluated the addition of nivolumab to adjuvant chemotherapy after upfront gastrectomy for gastric cancer (abstract 4000). A total of 755 patients with pathologic stage III curatively resected gastric cancer were randomized to nivolumab or placebo for 1 year plus adjuvant chemotherapy: investigator's choice of 6 months of capecitabine/oxaliplatin (65% of patients) or 1 year of S-1 (35%). Overall, 86% of cancers were in the distal stomach, 56% had diffuse histology, and 11% were positive for tumor expression of PDL-1 (tumor proportion score [TPS], ≥1%).

There was no difference in the primary endpoint of centrally assessed 3-year relapse-free survival with nivolumab-chemotherapy compared with placebo-chemotherapy (65.3%; HR, 0.90; P=0.4363). No patient subset showed benefit from adjuvant nivolumab except for the small group with TPS-positive cancers.

Comment: This sobering negative result raises questions about the utility of immune checkpoint inhibitors as an adjuvant therapy in gastric cancer. Results from other recently completed trials of immune checkpoint inhibitors in the neoadjuvant and adjuvant setting are awaited.

Two trials evaluating HER2-targeted therapies in patients with previously chemotherapy-treated HER2+ advanced biliary tract cancers showed a high degree of response with novel agents.

The industry-funded, phase 2 HERIZON-BTC-01 trial evaluated zanidatamab, an antibody targeting two epitopes on HER2 (abstract 4008). Of the 80 patients, 51% had gallbladder primaries and 78% tested IHC 3+ for HER2. The response rate was 41.3% and the median duration of response was 12.9 months. The median progression-free survival (PFS) was 5.5 months. Toxicities, including grade 3/4 diarrhea (4.6%) and decline in cardiac ejection fraction (3.4%), were manageable.

The industry-funded phase 2 SGNTUC-019 trial evaluated the combination of tucatinib, an oral HER2-directed tyrosine kinase inhibitor, and trastuzumab (abstract 4007). Of the 30 patients with locally determined HER2+ status by IHC or gene amplification, 50% had gallbladder primaries. The response rate was 46.7% and the median duration of response was 6.0 months. The median PFS was 5.5 months and median overall survival was 15.5 months. Grade 3/4 toxicity, including diarrhea, was manageable.

Lastly, researchers presented updated results and quality-of-life data from the industry-sponsored, open-label, phase 3 IMbrave050 trial in hepatocellular cancer (abstract 4002). Patients at high risk for recurrence after resection or ablation with curative intent were randomized to observation alone or 1 year of adjuvant therapy with bevacizumab and atezolizumab.

Of the 668 patients, 62% had hepatitis B, 88% underwent tumor resection, and 90% had resection of a solitary tumor. At a median follow-up of 17.4 months, relapse-free survival, the primary endpoint, was improved with adjuvant therapy compared with observation (HR, 0.72; P=0.012); however, the curves came together at 21 months, resulting in similar median relapse-free survival in the two groups (22.1 and 21.4 months, respectively). There was no clinically meaningful deterioration in patient-reported quality of life (measured on the IL42-EORTC QLQ-C30 questionnaire) with adjuvant treatment compared with observation. Further follow-up from this trial is anticipated.