14th World Conference on Lung Cancer

Amsterdam, Netherlands

 

Abstract ID P2.167: Blind prediction of response to neoadjuvant treatment with erlotinib in early stage non-small cell lung cancer (NSCLC) using kinase activity profiles.

 

Type: Peer Review Abstract

Topic: 2. Translational Research / b. Prognostic and Predictive Biomarkers

Authors: R. Hilhorst1, E.E. Schaake2, R. van Pel2, P. Nederlof2, L. Houkes1, M. Mommersteeg1, R. de Wijn1, R. Ruijtenbeek1, M.M. van den heuvel2, P. Baas2, H. Klomp2; 1PamGene International BV 's-

Hertogenbosch/NETHERLANDS, 2NKI-AVL Amsterdam/NETHERLANDS

 

Background

It has been well established that subcategories of NSCLC patients may benefit from epidermal growth factor

receptor (EGFR) tyrosine kinase inhibitors (TKI’s), notably those with cancers that harbor mutated EGFR and wild type kRAS. However, certain patients lacking EGFR mutations might also benefit from TKI treatment. Previously (Hilhorst et al, J. Clin. Oncol. 28:7s, 2010 (suppl abstr 10566), we have shown that a classifier based on kinase activity profiles generated in the presence and absence of a kinase inhibitor can predict erlotinib response of NSCLC patients in a neoadjuvant setting. The aim of the current study was to evaluate this classifier using a blinded test set of patient tumor samples

 

Methods

Frozen tumor tissue was obtained from NSCLC patients (stage IA-IIIA) who were given 21 days of neoadjuvant

treatment with erlotinib prior to complete surgical resection. All specimens were analyzed for EGFR

and kRAS mutation status. Tissue cryosections were lysed in M-PER buffer supplemented with phosphatase

and protease inhibitors. Kinase activity profiles of the lysates were generated in the presence and absence of erlotinib on PamChip® microarrays comprising 144 tyrosine containing peptides derived from known human protein phosphorylation sites. Clinical response evaluation to erlotinib was based on histopathological examination of the tumor tissues and metabolic changes. A classifier for response prediction was established for this training set and subsequently applied to a blinded test set of 13 patients.

 

Results

Phosphorylation profiles were ATP dependent and differed between the patient samples. Addition of erlotinib

to the samples resulted in inhibition of signals. Based on these on-chip inhibition profiles, a PLS-DA (partial

least square discriminant analysis) classifier was obtained that distinguished responders and non-responders in the training set of 14 samples. Leave-one-out cross validation resulted in correct classification of 11 samples (79%). Application to the blinded test set of 13 samples resulted in correct prediction of outcome for 11 of 13 samples (85%). The training set contained one sample with an EGFR exon 19 mutation, one with an exon 20 in frame deletion and 2 samples with kRAS mutations. The test set contained two samples with exon 19 mutations and two samples with kRAS mutations. These results suggest that more patients might be eligible for erlotinb treatment.

 

Conclusion

This blinded study validates the use of a classifier based on kinase activity profiles of patients’ own tumor

tissue to predict the response to treatment. This test that measures drug effects at the molecular level, may be an important step towards personalized medicine