1195P - Detection of anaplastic lymphoma kinase (ALK) gene rearrangement in non-small cell lung cancer (NSCLC): a glance at “borderline” cases

Date 29 September 2012
Event ESMO Congress 2012
Session Poster presentation I
Topics Biomarkers
Non-Small-Cell Lung Cancer, Locally Advanced
Presenter Maximilian von Laffert
Authors M. von Laffert1, E. Berg1, D. Lenze1, P. Lohneis1, M. Hummel1, M. Dietel2
  • 1Institue Of Pathology Charité, Charite Berlin Mitte, 10117 - Berlin/DE
  • 2Surgical Pathology, Charité, Humboldt-Universität zu Berlin, Berlin/DE



Lung cancer is the leading cause of cancer related mortality in men and women. In ∼5% [range: 0.9-13] of NSCLC a genomic EML4-ALK fusion has been described, showing therapeutically sensitivity for ALK-inhibitors. Although much has been reported about clinical and histological data of this NSCLC subtype (young patients, non-smokers, solid/mucinous adenocarcinomas), the parameters for FISH-based (e.g. threshold of positive cells, signal distance) diagnosis vary among the different studies.

Material and methods

We performed retrospective screening (tissue microarray based) of 488 NSCLC (336 adenocarcinomas, 152 squamous cell carcinomas) for ALK-expression (IHC-intensity: 0-3) using immunohistochemistry (Novocastra) and ALK-breaks using FISH (Abbott, ALK break-apart with positivity if split signals or single red signals were detectable in ≥15% per 100 cells). The evaluation of the IHC and FISH data was performed independent from each other. Since various thresholds for ALK-break positivity have been reported in the literature, special attention was given to cases with ALK-breaks in the range of 12%-20%.


95.7% (467/488) of the cases were ALK-break-negative (<15%), 4.3% (21/488) ALK-break positive (≥15%). Concordant absence of ALK-expression (0) and ALK-breaks (<15%) was found in 71.1% (347/488), concordant ALK-expression (1-3) and ALK-breaks (≥15%) in 2.9% (14/488). Apart from the most commonly applied threshold for the percentage (15%) of ALK-break positive cells, we extended our view to the range between 12%-14% and 15%-20%: 41 cases displayed ALK-break positive cells in a range between 12% and 14%, with ALK-expression in 29.3% (12/41), 18 cases in the range between 15% and 20%, with ALK-expression in 61.1% (11/18). Only 3 cases showed more than 20% ALK-break positive cells all with ALK-expression.


The definition of ALK-positive cases eligible for therapy with ALK-inhibitors (e.g. Crizotinib) is – beyond clear-cut negative and positive cases – a challenge. There is a considerable number of “borderline” cases with 12%-20% ALK-break positive cells which needs special attention. Thus it is questionable which threshold should be finally applied to qualify patients for therapy with ALK-inhibitors and what might be the impact of ALK-expression for this decision.


All authors have declared no conflicts of interest.