Comparison of ASTRUM-006 and MATTERHORN Trials
Both ASTRUM-006 and MATTERHORN demonstrated significant event-free survival benefits with perioperative immunotherapy in resectable gastric/gastroesophageal junction cancer, but MATTERHORN provides stronger global evidence with broader applicability, while ASTRUM-006 offers a chemotherapy-sparing adjuvant approach specifically for PD-L1-positive patients.
Trial Design and Population
MATTERHORN Trial 1
- Design: Global, double-blind, placebo-controlled phase III trial
- Population: 948 patients (474 per arm) with resectable gastric/GEJ adenocarcinoma, unselected for PD-L1 status
- Geographic scope: Multinational (Western and Asian countries)
- Treatment: Durvalumab + FLOT (4 cycles perioperative) followed by durvalumab maintenance (10 cycles)
- Median follow-up: 31.5 months
ASTRUM-006 Trial 2
- Design: Double-blind, multicentre phase III trial
- Population: 588 patients (292 vs 296) with PD-L1 CPS ≥5 resectable gastric/GEJ adenocarcinoma
- Geographic scope: Primarily China (57 hospitals) plus Thailand
- Treatment: Serplulimab + SOX (3 cycles neoadjuvant) followed by serplulimab maintenance (17 cycles) vs SOX (3 cycles neoadjuvant + 5 cycles adjuvant)
- Median follow-up: 35.9 months (ITT population)
Efficacy Outcomes
Event-Free Survival
MATTERHORN 1:
- 2-year EFS: 67.4% vs 58.5% (HR 0.71,95% CI 0.58-0.86, P<0.001)
- Benefit observed across unselected population
ASTRUM-006 2:
- Median EFS in ITT (CPS ≥5): Not reached vs 35.9 months (HR 0.73,95% CI 0.56-0.94, P=0.015)
- Median EFS in CPS ≥10: Not reached vs 42.0 months (HR 0.65,95% CI 0.47-0.90, P=0.0082)
- Advantage: Greater benefit in higher PD-L1 expression subgroup
Overall Survival
MATTERHORN 1:
- 2-year OS: 75.7% vs 70.4%
- Piecewise HR: 0.99 (months 0-12) and 0.67 (≥12 months), P=0.03
- Weakness: Did not meet prespecified significance threshold (P<0.0001)
ASTRUM-006 2:
- OS data pending extended follow-up
- Weakness: Immature survival data at current analysis
Pathological Complete Response
MATTERHORN 1:
- pCR: 19.2% vs 7.2% (RR 2.69,95% CI 1.86-3.90)
- Advantage: Nearly 3-fold improvement
ASTRUM-006 2:
- pCR data not reported in available evidence
Safety Profile
MATTERHORN 1
- Grade 3-4 treatment-related AEs: 71.6% vs 71.2%
- Delayed surgery: 10.1% vs 10.8%
- Delayed adjuvant treatment: 2.3% vs 4.6%
- Advantage: No increase in perioperative complications with immunotherapy
ASTRUM-006 2
- Grade ≥3 treatment-related AEs: 47% vs 59%
- Treatment discontinuation due to AEs: 7% vs 11%
- Advantage: Lower toxicity profile, potentially due to chemotherapy-sparing adjuvant phase
Key Strengths and Weaknesses
MATTERHORN Strengths 1, 3, 4
- Global applicability: First multinational phase III evidence for perioperative immunotherapy in this setting
- Unselected population: No biomarker requirement increases accessibility
- Established chemotherapy backbone: FLOT is Western standard of care per FLOT4 trial 5
- Robust EFS benefit: Statistically significant primary endpoint
- Surgical feasibility: No increase in perioperative complications
MATTERHORN Weaknesses 1, 4
- OS data immature: Did not meet stringent prespecified significance threshold
- Intensive chemotherapy: FLOT regimen more toxic than Asian standards
- Treatment burden: 4 cycles perioperative chemotherapy plus 10 cycles immunotherapy maintenance
- No biomarker enrichment: Cannot identify patients most likely to benefit
ASTRUM-006 Strengths 2
- Biomarker-selected population: PD-L1 CPS ≥5 enrichment likely enhances benefit
- Superior safety profile: Lower grade ≥3 AEs (47% vs 59%)
- Chemotherapy-sparing adjuvant: Only immunotherapy maintenance after surgery reduces cumulative chemotherapy exposure
- Longer follow-up: 35.9 months provides more mature EFS data
- Greater EFS benefit in CPS ≥10: HR 0.65 suggests stronger effect in biomarker-high patients
ASTRUM-006 Weaknesses 2
- Geographic limitation: Primarily Chinese population limits generalizability
- Biomarker requirement: PD-L1 testing mandatory, excludes CPS <5 patients
- Non-standard chemotherapy backbone: SOX not established Western standard (though validated in Asian RESOLVE trial 5)
- OS data pending: Cannot confirm survival benefit yet
- Smaller sample size: 588 vs 948 patients provides less statistical power
Clinical Context and Implications
Guideline Perspective 5
Current guidelines weakly recommend perioperative chemotherapy or neoadjuvant chemoradiotherapy for resectable advanced esophagogastric junction cancer, with regional variations (Western: perioperative FLOT; Asian: surgery + adjuvant chemotherapy). Both trials advance this paradigm by integrating immunotherapy.
Comparative Trial Context 4
Critical distinction: KEYNOTE-585 (pembrolizumab + cisplatin/5-FU) failed to meet its primary EFS endpoint despite trends toward improvement, while MATTERHORN succeeded. This suggests the chemotherapy backbone and immunotherapy agent selection matter significantly.
Practical Decision Algorithm
For Western/International Practice:
- First choice: MATTERHORN regimen (durvalumab + FLOT) 1
- Rationale: Global evidence, unselected population, established FLOT backbone
- Use regardless of PD-L1 status
For Asian Practice:
- If PD-L1 CPS ≥5: Consider ASTRUM-006 regimen (serplulimab + SOX) 2
- Rationale: Superior safety, chemotherapy-sparing adjuvant, validated SOX backbone in Asia
- If PD-L1 CPS <5 or unknown: MATTERHORN approach remains option
- Alternative: Standard perioperative SOX per RESOLVE trial 5
Key Caveats:
- Both trials require extended OS follow-up to confirm survival benefit
- PD-L1 testing should be performed when feasible to identify patients most likely to benefit from ASTRUM-006 approach
- Neither trial addresses HER2-positive disease (requires trastuzumab integration per ToGA trial 5)
- Surgical quality and timing remain critical—neither regimen should delay appropriate surgery