Positron emission tomography (PET) as a predictive measure in patients with metastatic pancreatic cancer and normal CA19-9 levels at baseline.

Two recent MPACT subanalyses (Chiorean et al. and Ramanathan et al. Ann Oncol. 2016) demonstrated evidence that decreases from baseline in carbohydrate antigen 19-9 (CA19-9) and tumor uptake of radioactively labeled glucose (18F-FDG) as measured by positron emission tomography (PET) imaging were each significantly associated with longer overall survival (OS) in patients who received first-line treatment with nab-paclitaxel plus gemcitabine or gemcitabine for metastatic pancreatic cancer [1, 2]. These modalities are complementary approaches to monitor treatment efficacy in most patients. However, we raised the question of whether tumor response measured by PET could predict outcome for a subset of patients (15%–20%) with pancreatic cancer who do not secrete elevated levels of CA19-9 [3, 4]. In the MPACT trial, more patients experienced a metabolic response (MR) measured by PET imaging than a tumor response measured by computed tomography. PET imaging may be particularly valuable to predict outcomes in patients without elevated baseline CA19-9 levels.

We carried out a post hoc, pooled treatment-arm analysis of OS by PET response in patients without elevated CA19-9 levels at baseline (defined as <37 U/ml). Results for OS by MR (defined by European Organization for the Research and Treatment of Cancer [EORTC] criteria [5]) are summarized in Table 1. MRs at week 8 and at best response during the study were each significantly associated with longer OS in patients with no CA19-9 elevation at baseline. Specifically, the median OS in patients with an MR at week 8 was 5.2 months longer than that in patients without one. In addition, patients with an MR at any time during the study had a nearly twofold longer median OS than those without one.

Table 1.

Overall survival by metabolic response in patients without elevated baseline CA19-9 levels treated with nab-paclitaxel plus gemcitabine or gemcitabine alone (pooled)

	Patients without elevated CA19-9 levels at baseline in the PET cohort		HR (P-value)
	Yes PET MR	No PET MR
Week 8	n = 21	n = 17
Median OS, months	13.2	8.0	0.34 (0.001)
(95% CI)	(9.20–17.05)	(3.38–14.69)
One-year survival rate	52%	35%
Best response during treatment	n = 24	n = 16
Median OS, months	13.6	6.9	0.31 (0.003)
(95% CI)	(9.26–22.83)	(3.38–10.22)
One-year survival rate	58%	25%

CA19-9, carbohydrate antigen 19-9; MR, metabolic response; OS, overall survival; PET, positron emission tomography.

These results illustrate that PET imaging may serve as a valuable tool to monitor treatment response in the subset of patients without elevated CA19-9 levels at baseline. Future trials may examine this issue more systematically as a preplanned end point.

funding

This work was supported in part by a grant from Stand Up 2 Cancer and Celgene Corporation (no grant number).

disclosure

RKR: consultant or advisory role, honoraria, and research funding, Celgene Corporation; RK: research funding, Celgene Corporation; EGC: research funding, Celgene Corporation; HL: employment and stock ownership, Celgene Corporation; DDVH: consultant or advisory role, honoraria, and research funding, Celgene Corporation.