Abstract 1167P
Background
Previously, our group showed the potential use of a score (IPS) exclusively based on MGMT, NDRG-1 and PHLDA-3 immunohistochemistry (IHC) expression as strong prognosis factor in operated PanNET (Viúdez et al. Oncotarget 2016). The present retrospective multicenter project analyzes the predictive/prognosis role of MGMT, NDRG-1 and PHLDA-3 IHCs in patients with advanced PanNET treated CAPTEM or everolimus.
Methods
IHC nuclear staining for MGMT and PHLDA-3 were scored as 0, 1-5%, 6-50% and ≥ 51%. For NDRG-1, a cytoplasmic score from 0 to 2 based on distribution pattern (null, patched or diffuse) or intensity were established. mIPS was calculated based on those IHC expressions 0% IHC expression for MGMT: 0 pt; <50% PHLDA3: 0 pt; and NDRG1 pattern: 0-2 (total min: 0 pt, max: 4 pt).
Results
71 patients (pts) were included, 34 of them women (47.9%) and median age of 61.2 (20.1-81.2). 32 and 39 pts received CAPTEM or everolimus, respectively. Median previous lines were 1 (95.7% received somastotatin analogs-SA- previously), and 90.5% pts had ki67 <20%. 63.2%, 10.1% and 33.3% were null in MGMT, NDRG-1 and PHLDA-3 IHC observation. Significant number of patients received SA in combination with everolimus (p=0.011). In the entire cohort K-M analysis showed significant differences for PFS based on number of previous lines or disease control rate (p=0.05, p=0.0001, respectively). Significant differenced were also observed for OS based on number previous lines received (p=0.0001) IPS (p=0.001) or mIPS (p=0.0166). In pts treated with everolimus, multivariate Cox analysis showed significant differences for OS for SPI (HR: 22.40 CI95%: 1.78-280.91, p=0.016) and previous lines (HR: 0.067 CI95%: 0.06-0.74, p=0.028) based on our SPI. Similar results were observed in CAPTEM cohort for OS (p=0.011 HR: 0.065 CI95%: 0.008-0.541) and previous lines (p=0.002 HR: 0.011 CI95%: 0.001-0.186) according to original IPS.
Conclusions
From our knowledge it is the first time that a simple IHC score could be extremely useful to predict outcome in PanNET pts treated with everolimus or CAPTEM. More mature results and analyses of our project are needed and will be presented during 2020 ESMO Congress.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
GETNE.
Disclosure
All authors have declared no conflicts of interest.