How Should Response to Neoadjuvant Therapy Be Assessed in Pancreatic Cancer?: Annals of Surgery | August 2026

Introduction:

Neoadjuvant therapy (NAT) has become an integral component of treatment for patients with operable pancreatic ductal adenocarcinoma (PDAC). However, assessing treatment response remains challenging because biochemical markers (CA19-9), radiographic imaging, and pathological tumor regression often provide conflicting results. This study compared these three response assessment methods to determine which best predicts long-term survival after pancreatic resection.

Why was this study needed?

There is no consensus on the best method to assess response after neoadjuvant therapy in PDAC.

Radiological tumor shrinkage often underestimates treatment response because of persistent fibrosis.

CA19-9 reduction is widely used but may not fully reflect tumor biology.

Pathological tumor regression is informative but is only available after surgery.

Identifying the most reliable predictor of survival could improve treatment decisions and postoperative risk stratification.

Results:

The study included 225 patients who underwent pancreatic resection after neoadjuvant therapy, most commonly FOLFIRINOX.

A greater than 50% reduction in CA19-9 after neoadjuvant therapy was associated with significantly improved overall survival, making biochemical response a stronger predictor than radiological response.

Radiographic tumor shrinkage was not associated with improved survival, highlighting the limitations of imaging alone in evaluating treatment response.

Pathological tumor regression alone also showed limited ability to predict long-term survival.

The best survival outcomes were observed in the small subgroup of patients who achieved both a major CA19-9 response and major pathological tumor regression, with a median overall survival of approximately 40 months.

This favorable dual-response pattern was more frequently achieved when neoadjuvant radiotherapy was added after systemic chemotherapy, suggesting that combined-modality treatment may enhance deep tumor response in selected patients.

Clinical Impact:

This study demonstrates that radiological response should not be the primary measure of treatment success in pancreatic cancer after neoadjuvant therapy. Instead, CA19-9 reduction provides a more meaningful indicator of prognosis, while the combination of biochemical and pathological response identifies patients with the most favorable long-term outcomes. These findings support a multimodal approach to response assessment and reinforce the importance of integrating tumor biomarkers with pathology rather than relying solely on imaging.

Bottom Line:

In resected pancreatic cancer, CA19-9 response is a better predictor of survival than radiographic tumor shrinkage after neoadjuvant therapy. Patients achieving both significant CA19-9 reduction and major pathological tumor regression have the best long-term survival, supporting a combined biochemical and pathological approach to treatment response assessment.