T-DXd 5.4 mg/kg Combos Yield Fewer Grade ≥3 AEs in HER2+ Gastric Cancers

Regimens combining fam-trastuzumab deruxtecan-nxki (T-DXd; Enhertu) at 5.4 mg/kg with an immune checkpoint inhibitor (ICI) and a lower-dose fluoropyrimidine produced lower rates of toxicity compared with regimens utilizing T-DXd at 6.4 mg/kg and higher-dose fluoropyrimidine among patients with advanced HER2-expressing gastric cancer (GC), gastroesophageal junction adenocarcinoma (GEJ), or esophageal adenocarcinoma (EA), according to safety findings from the phase 1b/2 DESTINY-Gastric03 trial (NCT04379596). These results from part 2—arms D and F—and part 4 of the trial were presented at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting.

Treatment and Dosage Across the Trial

1. Part 2 arm D (n = 43): T-DXd at 6.4 mg/kg plus higher-dose fluoropyrimidine and pembrolizumab (Keytruda)
2. Part 2 arm F (n = 32): T-DXd at 5.4 mg/kg plus lower-dose fluoropyrimidine and pembrolizumab
3. Part 4 (n = 62): T-DXd at 5.4 mg/kg plus lower-dose fluoropyrimidine and rilvegostomig

Safety Findings

Treatment-related grade 3 or higher AEs occurred in 76.7% of patients in part 2 arm D, compared with 34.4% of patients in part 2 arm F, and 38.7% of patients in part 4. Serious AEs occurred in 58.1%, 37.5%, and 35.5% of patients in parts 2 Arm D, 2 Arm F, and 4, respectively; treatment-related serious AEs occurred in 44.2%, 15.6%, and 19.4% across the 3 groups.

“Rates of [grade 3 or higher] all-cause and treatment-related AEs were numerically lower among patients receiving T-DXd 5.4 mg/kg in combination with an [immune checkpoint inhibitor] and a lower-dose fluoropyrimidine than in those treated with a regimen containing T-DXd 6.4 mg/kg and a higher-dose fluoropyrimidine,” wrote lead study author Yelena Y. Janjigian, MD, Carroll and Milton Petrie Chair and chief of Gastrointestinal Oncology Service at Memorial Sloan Kettering Cancer Center, and coauthors, in the presentation.

Per adverse events of special interest (AESIs) related to T-DXd, the most clinically meaningful difference in the safety comparison between groups was the rate of adjudicated treatment-related interstitial lung disease (ILD) or pneumonitis. In part 2 arm D, ILD occurred in 6 patients (14.0%), including 2 grade 5 events and 3 grade 2 events. By contrast, no ILD events were observed in part 2 arm F, and 2 patients (3.2%) in part 4 experienced ILD, both grade 2 in severity. Left ventricular dysfunction was observed in 1 patient (1.6%) in part 4, with no events in part 2 arms D or F.

AEs leading to discontinuation of any investigational product were also lower in the 5.4 mg/kg arms. Discontinuations of T-DXd occurred in 23.3%, 6.3%, and 3.2% of patients in part 2 arm D, part 2 arm F, and part 4, respectively. AEs led to discontinuation of pembrolizumab in 20.9% and 6.3% of patients in parts 2 arm D and part 2 arm F, respectively; discontinuation of rivostogimab occurred in 3.2% of patients in part 4. Rates of AEs leading to T-DXd dose reduction were 30.2%, 15.6%, and 21.0% across the 3 groups.

AEs associated with death were reported in 4 patients (9.3%) in part 2 arm D and 2 patients (3.2%) in part 4, with no instances in part 2 arm F.

Across all 3 groups, the most frequently reported any-grade treatment-related AE was nausea, occurring in 24 patients (55.8%) in part 2 arm D, 13 patients (40.6%) in part 2 arm F, and 34 patients (54.8%) in part 4.

The most common grade 3 or higher treatment-related AEs in part 2 arm D were anemia (20.9%), nausea (16.3%), diarrhea (14.0%), aspartate aminotransferase elevation, (11.6%) thrombocytopenia, and hypokalemia (11.6%). In part 2 arm F, the most common grade 3 or higher events were neutrophil count decreased (12.5%) and anemia (6.3%). In part 4, grade 3 or higher events were led by neutrophil count decreased (11.3%) and decreased appetite (6.5%).

The most common AESIs related to rivostogimab in part 4 were diarrhea/colitis and hepatic events, each occurring in 9 patients (14.5%), with most events grade 1 or 2 in severity.

“The safety profiles of all 3 regimens were generally consistent with those of each component agent, with no new safety signals identified,” the authors added.

How is the DESTINY-Gastric03 trial designed?

DESTINY-Gastric03 is a global, multicenter, open-label, dose-escalation and expansion trial evaluating T-DXd-based combination regimens in patients with HER2-expressing advanced GC, GEJA, or EA with no prior treatment for metastatic disease. The study enrolled patients aged 18 years or older with an ECOG performance status of 0 or 1 and HER2-expressing unresectable, locally advanced, or metastatic disease. The median treatment duration was 6.1 months for part 2 arm D (data cutoff, February 15, 2023), 6.9 months for T-DXd in part 2 arm F (data cutoff, August 19, 2024), and 6.5 months in part 4 (data cutoff, October 16, 2025). Notably, the data cutoff for each arm was selected to maintain similar treatment durations

The primary end point of the study was confirmed overall response rate (ORR) by investigator assessment. Secondary end points included safety, disease control rate, survival, and pharmacokinetics.

“These results support the continued investigation of T-DXd 5.4 mg/kg in combination with pembrolizumab or rilvegostomig as [first line] treatment in patients with advanced HER2-expressing GC, GEJA, or EA,” concluded the study authors.

References

Janjigian YY, Shoji H, Kowalczyk A, et al. First-line trastuzumab deruxtecan (T-DXd)–based regimens in advanced HER2-expressing gastric cancer, gastroesophageal junction adenocarcinoma, or esophageal adenocarcinoma: Safety results from DESTINY-Gastric03 part 2 arms D and F, and part 4. J Clin Oncol. 2026;44(suppl 16):4022. doi:10.1200/JCO.2026.44.16_suppl.4022