Molecular pathology before systemic treatment: feasible?

Because high-throughput sequencing techniques have entered the molecular landscape of metastatic prostate cancer, we can start implementing precision oncology in the treatment of patients with advanced prostate cancer². Identifying prognostic tumour characteristics as well as predictive biomarkers will help guide clinical management. Molecular pathology plays an important role in establishing these biomarkers in tumour tissue or liquid biopsy. However, we should acknowledge that most patients are likely treated outside of large, mainly university-based treatment centres specialising in urological oncology. Therefore, access to very novel and in general cost-intensive molecular studies may not always be available. We will hence focus on biomarkers that have the potential of being broadly implemented in everyday practice and on techniques that are already established. Reimbursement cannot be part of our considerations due to regional differences.

Immunohistochemistry
The most used adjunct diagnostic method in virtually all pathology labs is immunohistochemistry. Immunohistochemistry is a reliable method to visualise proteins on a cellular level. Both qualitative (is the protein expressed?) and (semi-) quantitative (how much of the protein is expressed and in which cellular compartment?) immunohistochemistry is available. The latter is especially important in the setting of predictive biomarkers, e.g. PD-L1- or HER2-expression on tumour cells.

Another method commonly available in most pathology labs is fluorescence-in-situ-hybridization (FISH) to assess numeric or structural chromosomal alterations. It is commonly used in non-small-cell lung cancer (NSCLC) to detect therapeutic targets such as ALK or ROS1 rearrangements. Next-generation sequencing methods were pioneered in NSCLC as well. Most labs nowadays perform either in-house targeted sequencing or send their samples for sequencing to external collaboration partners. Whole-exome or whole-genome-sequencing is uncommon in daily practice. Liquid biopsies interrogating treatment resistance to EGFR-TKI are more widely implemented as well and may be adapted to serve the needs of prostate cancer patients. This is particularly the case in tumour heterogeneity that might not be sufficiently addressed by a biopsy of a single metastatic site². Which pathways can therefore be interrogated before starting systemic treatment in patients with advanced and metastatic prostate cancer?

AR pathway
The androgen receptor (AR) pathway is the most altered pathway in prostate cancer with a variety of alterations such as AR amplification, mutation, and chromosomal rearrangements. AR overexpression and the presence of several splice variants are frequent in advanced and metastatic prostate cancer. AR aberrations detected in cell-free DNA via liquid biopsies can provide valuable information about resistance to androgen inhibitor therapy, such as enzalutamide and abiraterone³. Detection of the L702H mutation both in cell-free DNA and circulating tumour cells (CTCs) is associated with resistance to AR inhibitor drugs^4,5. Linking the detection of the AR splice variant AR-V7 in CTCs to shorter progression free and overall survival has gained a lot of interest as predictive biomarker for abiraterone and enzalutamide therapy^6,7. However, since some patients might still respond to abiraterone and/or enzalutamide, these tests have not been implemented in routine practice yet⁸.

PI3K-AKT-pathway
Major alterations are also found in the PTEN-PI3KAKT-axis. Phosphatase and tensin homologue (PTEN) loss, mostly by deletion, is frequent in advanced prostate cancer and can be detected by immunohistochemistry or FISH analysis^9,10. Loss of PTEN is found in up to 50% of CRPC. PTEN is a negative regulator of the PI3K-AKT-pathway and upon deletion, the PI3K-AKT-pathway is activated resulting in increased surrvival¹¹. In CRPC samples, PTEN loss might be predictive of decreased response to abiraterone and of therapeutic response to PI3K inhibitor therapy^11,12. However, given the intricate mechanism governing both the PI3K-AKT-pathway and the AR pathway and their crosstalk, a promising PI3K-inhibitor therapy for patients with advanced prostate cancer has not yet emerged¹¹. PTEN loss is also a strong prognostic biomarker both in primary prostate cancer and in CRPC. In CRPC, PTEN loss is associated with shorter survival and shorter time on AR-inhibitor therapy¹³.

DNA damage repair pathway
Defects in the DNA damage repair (DDR) pathway are much more common than previously thought. In patients with advanced or metastatic prostate cancer, the frequency of germline DDR ranges between 11 and 15%^14,15. Overall, the frequency of DDR is about 20-25% in CRPC². Both germline and somatic DNA damage repair deficiencies are predictive of response to PARP-inhibitor therapy¹⁶. Next-generation sequencing (NGS) to detect DDR is commonly available. It should be noted that pathogenic mutations in BRCA1/2 are spread widely across both genes and NGS interrogating DDR should also include genes such as ATM or CHEK2. Defects in the DNA damage repair pathway have also been associated with exceptional response to platinum-based chemotherapy¹⁷. According to the 2017 advanced prostate cancer consensus conference (APCCC), BRCA1/2 and ATM mutations should be analysed and reported if a patient undergoes metastatic biopsy.

“Novel prognostic and predictive biomarkers keep emerging as we enhance our understanding of the molecular basis of advanced and metastatic prostate cancer.”

Repair genes
Some patients harbour a defect in one of the mismatch repair genes (MSH2, MSH6, PMS2, MLH1), resulting in a hypermutated phenotype and microsatellite instability (MSI). Their disease may be amenable to immunotherapy. Pembrolizumab has been approved for patients harbouring a defect in the genes mentioned above, regardless of underlying histology. The MSI status can be easily assessed by immunohistochemistry and followed up by molecular studies, if immunohistochemistry detects a loss of the mismatch repair proteins².

Neuroendocrine prostate cancer
Lineage plasticity with transformation from an adenocarcinoma phenotype to a highly aggressive, AR-independent neuroendocrine phenotype via TP53 and RB1 alterations remains a clinical challenge⁸. If such a transformation is suspected, a metastatic biopsy might be indicated to rule out or confirm small cell morphology. If a transformation occurred, a routine HE stain without necessity for further molecular analyses is enough. In small cell carcinoma, the Notch signalling pathway with overexpression of DLL3 is active. If small cell carcinoma is present, assessing protein expression of DLL3 might serve as a predictive biomarker for anti-DLL3-therapy¹⁸.

In conclusion, novel prognostic and predictive biomarkers keep emerging as we enhance our understanding of the molecular basis of advanced and metastatic prostate cancer. Most of these biomarkers could be implemented in routine practice if requested by clinicians to guide patient management. For the time being, however, most of these predictive biomarkers are not sufficiently validated.

The reference list can be made available to interested readers upon request by sending an email to: communications@uroweb.org.