56P - Pro-GRP in small cell lung cancer

Date 07 May 2017
Event ELCC 2017
Session Poster Display Session
Topics Small Cell Lung Cancer
Thoracic Malignancies
Presenter Stefano Cavalieri
Citation Annals of Oncology (2017) 28 (suppl_2): ii17-ii20. 10.1093/annonc/mdx088
Authors S. Cavalieri1, F. Nichetti1, D. Morelli2, F. de Braud3, A. Martinetti1, E. Sottotetti1, K. Dotti1, M. Prisciandaro1, F. Corti1, M. Platania1
  • 1Medical Oncology Department, Fondazione IRCCS - Istituto Nazionale dei Tumori, 20133 - Milano/IT
  • 2Laboratory Department, Fondazione IRCCS - Istituto Nazionale dei Tumori, 20133 - Milano/IT
  • 3Medical Oncology Department - University Of Milan, Fondazione IRCCS - Istituto Nazionale dei Tumori, 20133 - Milano/IT



NSE was historically the recommended tumor marker for SCLC. It stains up to 80% of SCLCs in tissue examinations but is also elevated in the sera of 20–30% of patients with NSCLC. NSE has low sensitivity, particularly in patients with limited disease (LD). Recently, pro-gastrin-releasing peptide (pro-GRP) became available as a sensitive, specific and reliable tumor marker for patients with SCLC.


We retrospectively analyzed pro-GRP in pts with SCLC and NSCLC treated from 2015 to 2016 at our Center. Serum pro-GRP level was measured with electrochemiluminescence method at our laboratory; it was considered elevated if higher than 77.8 pg/ml. Statistical tests used to study differences between pro-GRP medians were Mann-Whitney U test for SCLC vs NSCLC and Fisher’s exact test for DD vs LD. To calculate sensitivity (Sn), specificity (Sp), positive and negative predictive values (PPV and NPV respectively) we considered: SCLC with elevated pro-GRP as true positive (TP), NSCLC with normal pro-GRP as true negative (TN), SCLC with normal pro-GRP as false negative (FN), NSCLC with elevated pro-GRP as false positive (FP).


A total of 46 pts were studied (33 men, 13 women), whose median age was 67 years (range 20-78). 20 pts had SCLC (16 DD, 4 LD) and 26 metastatic or unresectable advanced NSCLC. Median pro-GRP level was 1398 pg/ml (range 43-50000) in SCLC and 26 pg/ml (range 6-119.2) in NSCLC (difference 1372 pg/ml; SCLC/NSCLC ratio 53.76; p < 0.01). TPs were 18, FNs 2, TNs 25, FP 1. Sn was 90%, Sp 96.15%, PPV 94.73% and NPV 92.59%. All the TPs had diffuse disease, the 2 FNs had limited disease, the only FP had a poorly differentiated adenocarcinoma. Median pro-GRP was 182.25 pg/ml (range 43-642) in LD pts and 2522 pg/ml (range 88-50000) in those with DD (absolute difference 2339.75 pg/ml; ratio DD/LD 13.83; p = 0.026). Pro-GRP was elevated in all DD pts and in half of LD pts (2/4): Both of them had a high burden of loco-regional disease.


Our data confirm that pro-GRP is sensitive for SCLC diagnosis and that it might help to discriminate DD from LD. Since high marker levels are related to a high disease burden, pro-GRP may have a negative prognostic significance. Follow up is required to define its role in clinical practice in monitoring response to treatment.

Clinical trial identification

Not applicable

Legal entity responsible for the study

Fondazione IRCCS Istituto Nazionale dei Tumori


Fondazione IRCCS Istituto Nazionale dei Tumori


All authors have declared no conflicts of interest.