Six-months CA19.9 velocity as a predictive factor for relapse-free survival after curative pancreasectomy in pancreatic cancer

Surgery represents the only potentially curative therapeutic option for patients (pts) with pancreatic cancer, but it is burdened by a high rate of both local and systemic relapse, even with the use of adjuvant chemotherapy. Biomarkers for predicting cancer recurrence are still lacking. Carbohydrate antigen 19-9 (Ca19.9) currently represents the only commonly used marker, but there is limited evidence for the role of its kinetics during follow-up.

We retrospectively enrolled pts who carried out follow-up assessment after pancreatic surgery with radical intent at our center between 2015 and 2020. Patients and tumor characteristics and serum CA19.9 values were collected during the follow-up period. We analyzed the six-months CA19.9 velocity, defined as units of increase per milliliter every 3 months (U/mL/3 months). Aim of the study was to correlate six-months CA19.9 velocity with Relapse-Free Survival (RFS), one-year RFS and two-years RFS.

Of 226 enrolled patients, 115 were female and 111 were male; median age was 67 (range 44-89) years old. Most of the tumors involved the head (72%), whereas 28% body-tail. At pre surgery evaluation, pts with borderline resectable disease on CT scan were 36%, whereas in 64% there was no vascular involvement (VI). Pathological lymph node status (pN) was pN1 in 49%, pN2 in 27% of pts and pN0 in 24%. Resection margin status was positive in 37% and negative in 63%. Poorly-differentiated cancers represented 25% of cases, moderately differentiated were 44% of cases, and for the remaining cases grading was unknown. Median post-surgery CA19.9 level was 18,80 (0.60-36,474) U/mL. Adjuvant chemotherapy was performed in 66% of pts. At a median follow-up of 65 months, median overall-survival was 36 (2-98) months and median RFS was 15.9 (0-94,8) months. One-year RFS was 60% and two-year RFS was 40%. The correlation coefficient between six-months CA19.9 velocity in relapsed patients and RFS was r -0,2922 (P=0,0001). High six-months CA19.9 velocity, defined as >1 U/mL/3 months, predicted one-year RFS (Chi-squared 6,052; contingency coefficient 0,186; P=0,0139) and two-year RFS (Chi-squared 12,532; contingency coefficient 0,263; P=0,0004). Positive six-months CA19.9 velocity was a prognostic factor at univariate analysis, as Eastern Cooperative Oncology Group performance status (ECOG PS), tumor grading, adjuvant chemotherapy, post-surgery CA19.19 higher than 40 U/mL, pN, VI and positive margin (p < 0.05). At a multivariate analysis, six-months CA19.9 velocity confirmed its independent prognostic role for RFS (HR 1.51 95% CI 1.01-2.25, p=0.04). Post-surgery CA19.19 higher than 40 U/mL (HR 2,40 CI 1,6049 to 3,6088, p < 0.0001), adjuvant chemotherapy (HR 0.43 CI 0.2483 to 0.7511, p < 0.002), pN (HR 1.69 CI 1.2316 to 2.3256, p=0.001) and VI (HR 1.75 CI 1.727 to 2.627, p=0.006) remained as prognostic factors at multivariate analysis. Positive six-months CA19.9 velocity predicts radiographic recurrence with median time of 3,4 (range 0-22.6) months in pts whose recurrence occurs within the first 2 years.

Six-months CA19.9 velocity is a promising tool for predicting microscopic residual disease despite adjuvant chemotherapy, and its increase is able to predict in advance radiologic progressive disease.

The authors.

Has not received any funding.

All authors have declared no conflicts of interest.