Why this matters
Peritoneal metastases from colorectal cancer remain difficult to treat.
Radiographic assessment can underestimate the true extent of peritoneal disease, response to second-line systemic therapy is often limited, and only selected patients ultimately become candidates for cytoreductive surgery.
TORCH-PM01 tested an unusual conversion-oriented strategy combining three components:
- FAPI PET/CT for improved visualization of peritoneal disease
- hypofractionated radiotherapy to FAPI-defined targets
- cadonilimab added to second-line systemic therapy
The early randomized signal is substantial. Objective response increased markedly, more patients underwent cytoreductive surgery, and progression-free survival strongly favored the experimental strategy.
But this is a small phase II study, and the multimodality design makes it impossible to determine which component drove the benefit.
Study design
TORCH-PM01 was a prospective randomized phase II study in patients with colorectal cancer and peritoneal metastases progressing after first-line systemic therapy.
Forty patients were randomized to:
- standard second-line systemic therapy
- second-line systemic therapy plus cadonilimab and FAPI PET/CT-guided hypofractionated radiotherapy
Radiotherapy was delivered as:
- 20-25 Gy in 5 fractions
Cadonilimab was administered at 6 mg/kg every 2 weeks for up to 8 cycles.
Systemic therapy was investigator-selected in both groups.
Patients underwent imaging assessment every 8 weeks, with multidisciplinary review of potential resectability.
The primary endpoint was objective response rate.
Secondary and exploratory endpoints included:
- progression-free survival
- conversion to cytoreductive surgery
- peritoneal cancer index reduction
- pathological response
- safety
Twenty patients were randomized to each arm.
Key results
The primary endpoint strongly favored the experimental strategy.
Objective response in the intention-to-treat population was:
- 60% with FAPI-guided RT plus cadonilimab
- 15% with systemic therapy alone
- p=0.003
This represents an absolute improvement of 45 percentage points.
Among evaluable patients, the response rates were approximately 63% vs 16%, consistent with the intention-to-treat analysis.
Response depth was also greater.
Median reduction in measurable disease was approximately:
- 55.6% in the experimental arm
- 7.0% in the control arm
Complete responses occurred in approximately 10% of patients receiving the experimental strategy.
Peritoneal downstaging
The treatment effect was also visible in the peritoneal disease burden.
Median change in peritoneal cancer index was:
- -1.0 with the experimental strategy
- 0 with systemic therapy alone
- p=0.001
Any reduction in PCI occurred in:
- 57.9% with the experimental treatment
- 10.5% with control
- p=0.002
These findings are particularly relevant because the treatment strategy was designed around conversion to potentially definitive local therapy.
Conversion to surgery
More patients receiving the experimental regimen became candidates for cytoreductive surgery.
Conversion to CRS occurred in:
- 60% with FAPI-guided RT plus cadonilimab
- 20% with systemic therapy alone
R0 or R1 resection was achieved in:
- 50%
- 15%
Pathological complete response was observed in 3 patients in the experimental group and none in the control group.
These surgical outcomes are clinically interesting, but they were exploratory and depended on multidisciplinary assessment of resectability.
Progression-free survival
The presentation also showed a strong early PFS signal.
Median PFS was:
- not reached with the experimental strategy
- 7.9 months with systemic therapy alone
The estimated hazard ratio was:
- HR 0.20
- 95% CI 0.08-0.51
- p=0.0014
Six-month PFS was:
- 94.7% vs 73.7%
At 12 months:
- 58.6% vs 14.8%
This is one of the most compelling parts of the dataset, but follow-up remains short and overall survival data are not mature.
Toxicity
The more intensive treatment strategy produced more toxicity.
Grade 3-4 treatment-related adverse events occurred in approximately:
- 60% with the experimental regimen
- 45% with systemic therapy alone
Selected severe toxicities included:
- neutropenia
- nausea
- vomiting
- diarrhea
Immune-related adverse events were also observed with cadonilimab.
No new safety signal was identified, but the treatment burden is clearly greater than with systemic therapy alone.
Interpretation
TORCH-PM01 provides an intriguing proof-of-concept for treating colorectal peritoneal metastases as a conversion problem rather than simply a systemic disease problem.
The magnitude of response is notable.
Objective response increased from 15% to 60%, while conversion to cytoreductive surgery increased from 20% to 60%.
The PFS curves also separate early and substantially.
The FAPI component is particularly interesting because peritoneal metastases can be difficult to define anatomically on conventional imaging. FAPI PET/CT may improve target identification and allow radiotherapy to address disease that is otherwise poorly visualized.
But the study does not tell us whether FAPI-guided RT itself is responsible for the improvement.
The experimental intervention simultaneously introduced:
- FAPI-based imaging
- hypofractionated RT
- cadonilimab
- systemic therapy
The control arm received systemic therapy alone.
Therefore, the contribution of radiotherapy, immunotherapy, their interaction, and improved target visualization cannot be separated.
This is an important distinction for radiation oncology.
The study supports the multimodality strategy as a whole. It does not establish that FAPI-guided radiotherapy alone improves response or survival.
The surgical conversion data are also promising but should be interpreted cautiously. Resectability was determined by multidisciplinary review, and surgery-related endpoints were exploratory.
Limitations
The trial included only 40 randomized patients.
It was conducted at a single center.
The experimental arm combined several interventions simultaneously, preventing attribution of benefit to any individual treatment component.
Systemic treatment was investigator-selected, introducing heterogeneity in the treatment backbone.
Conversion to surgery was an exploratory endpoint and depended partly on multidisciplinary assessment and feasibility.
Follow-up is still short.
Overall survival is immature.
Secondary and exploratory analyses were not adjusted for multiple comparisons.