Correlation of elevated CA19-9 and cancer associated macrophages in blood with outcomes in pancreatic cancer patients over 5 years.

e16464 Background: Pancreatic cancer (PC) is a highly aggressive cancer accounting for ~7% of all cancer related deaths in the United States. In PC, Carbohydrate antigen 19-9 (CA19-9) is blood-based protein biomarker which linearly correlates with disease stage and at ≥37 U/mL is associated with worse clinical outcomes, though its clinical utilization is limited in ~20% PC patients (pts) with low/no CA19-9 secreting tumors. Recently, a phagocytic tumor macrophage found in the blood in parallel with CA19-9 (e.g. cancer-associated macrophage-like cells [CAMLs]), appears to be independently associated with poorer prognosis in PC, when engorged ≥50 µm. However, the relationship between CAMLs & CA19-9 remains unexplored. In this pilot study, we recruited n = 68 PC pts with clinically resectable or borderline resectable PC, to assess ≥50 µm CAMLs & ≥37 U/mL CA19-9 against progression free survival (PFS) and overall survival (OS) over 5 years. Methods: We initiated a prospective pilot study to procure blood samples from pts (n = 68) referred for surgical resection with clinically staged resectable (n = 50) or borderline resectable (n = 18) PC. Anonymized pre-surgical blood samples (7.5mL) were processed via CellSieve microfiltration, isolated, and enumerated for number of CAMLs ≥50 µm. In parallel, CA19-9 levels ≥37 U/mL were measured. The association of CA19-9 & CAMLs against PFS & OS was analyzed at 60 months, with hazard ratios (HRs) calculated by censored univariate and multivariate analyses. Results: CA19-9 was available for 96% (n = 65/68) of pts and CAMLs available for 84% (n = 57/68), with (n = 4) missing a pre-surgical blood draw, (n = 5) failing the CAML assay, and (n = 2) receiving treatment offsite. 55 pts had both CA19-9 and CAML results available. 48% (n = 31/65) of pts had CA19-9 ≥37 U/mL which was not significant for PFS (HR = 1.6, 95%CI = 0.8-2.9, p = 0.2230), but was significant for worse OS (HR = 2.0, 95%CI = 1.0-3.8, p = 0.0486). 68% (n = 39/57) of pts had CAMLs ≥50 µm which was significant for worse PFS (HR = 3.5, 95%CI = 1.8-6.7, p = 0.0004) and OS (HR = 3.7, 95%CI = 1.9-7.3, p = 0.0003). 40% (n = 22/55) of pts had both CA19-9 ≥37 U/mL & CAMLs ≥50 µm which was significant for worse PFS (HR = 5.0, 95%CI = 2.2-11.3, p = 0.0002) & OS (HR = 7.9, 95%CI = 3.4-18.2, p < 0.0001). At post-surgical pathological assessment it was found that the percentage of pts with locally advanced (LA) or metastatic PC (mPC) disease was, 86% when CAMLs ≥50 µm & CA19-9 ≥37 U/mL, 71% when CAMLs ≥50 µm but CA19-9 < 37 U/mL, 25% when CAMLs < 50 µm but CA19-9 ≥37 U/mL, or 25% when CAMLs < 50 µm & CA19-9 < 37 U/mL. Conclusions: The combination of CAMLs & CA19-9 shows promise as a non-invasive biomarker for stratifying pts with worse clinical outcomes and possibly identifying resectable pts with more advanced PC. Given these results, future studies with larger cohorts should evaluate and refine any potential use in guiding treatment decisions in PC.