‘Today, cancer therapy is experiencing a shift towards targeted medicine, but the predictive power of tissue based analyses is still underestimated,’ declared Professor Manfred Dietel, Director of Institute of Pathology at Charité Medical School, Berlin. Currently, routine detection of a predictive biomarker is a crucial prerequisite in the individualised or targeted treatment of the three ‘killer’ carcinomas (around 35% of all tumours): cancers of the breast, colon (metastasised, first and second line) and non-small cell lung cancer (stage IIIB and IV). Further, predictive analyses are common in the treatment of the more seldom gastrointestinal stromal-tumour (dose determination) and several types of lymphomas.
Targeted therapy rather started with the agent Trastuzumab, used against metastatic breast cancer. The agent’s FDA approval was gained in 1998 and EU approval came in 2000. ‘The development of a monoclonal antibody with the trade name Herceptin, by Roche, was the big breakthrough,’ the professor said.
The monoclonal antibody interferes with the HER2 receptor, but will only have therapeutic effects if an amplification of the HER2/neu gene or an over-expression of its protein can be verified in advance. Therefore early and accurate HER2 testing of all breast cancer patients at primary diagnosis is essential for optimal disease management.
This eligibility test is realised by morphologic and cytologic examinations. ‘We can show in one section of tissue if there is an amplification of Her2 gene or over-expression of protein. That assures a relatively high dependability,’ Prof. Dietel said, though admitting: ‘Nevertheless, in approximately 25-30% of the cases we can have the problem of adequate quantification due to borderline staining.’
This challenge is faced by Prof. Dietel and his Charité team by using virtual microscopy to scan histological slices, which then can be scaled up very easy and hence the problem of quantification corrected. ‘This method is still not routine, but in two or three years I think we will be able to verify the important molecules in the tissue with such instruments in an objective way.’
In the treatment against invasive colorectal cancer, predictive examinations are also inevitable. Two new substances have been introduced to clinical practice, Cetuximab and Panitumumab, both antibodies that attack the EGFR-receptor. But they will only operate if the tumour cells have a wild-type K-RAS gene (mutation excluded).
Whereas, in non-small cell lung cancer (NSCLC), targeted therapy proved effective only if mutation of the EGFR-receptor is ensured by molecular pathology. In April 2009 European Medicines Agency, the European FDA, gave only a conditional approval to Iressa (Gefitinib), a drug to treat NSCLC. ‘This was the first time, worldwide, that a drug approval was definitively associated with an eligibility test,’ Prof. Dietel explained, and again emphasised the necessity of pathological tissue-based examinations. ‘We need tissue-based examinations for almost all predictive eligibility tests. Whether we can detect gene mutations in a non-invasive way, as indicated by developments in molecular imaging, time will tell.’