Clinical Significance of AR-V567es in Prostate Cancer—Letter

There are conflicting results regarding incidence of ARv567es detected by PCR in prostate cancer, displaying frequencies ranging from 3% in metastatic to more than 70% in circulating tumor cells (CTC; refs. 1, 2). To resolve this issue, Bernemann and colleagues performed a comprehensive comparison of the different reported PCR detection strategies for ARv567es (1). Optical readout of fluorescence signals is the basis of SYBR Green and TaqMan approaches that permit quantitative measurements of the total PCR amplicons during the exponential stages. However, when primers are not well designed, PCR is associated with the presence of primer dimer and mis-priming product, leading to misinterpretation of the data derived from amplicon-based fluorescence signals (i.e., false-positive signals). Many published assays are based on PCR with two oligonucleotides, one of which spans the exon junction of exon 4 to exon 8 (2, 3). However, sequence similarity exists between the 5′ edges of exon 4 and exon 7, as well as the 3′ edges of exon 5 and exon 8. Thus, Bernemann and colleagues demonstrated nonspecific binding of exon spanning oligonucleotides might occur with AR-FL in the absence of ARv567es, providing false-positive signals for both SYBR Green- and TaqMan-based assays. Furthermore, they have shown that previously reported PCR primers and detection methods resulted in ARv567es detection with high false-positive rates in cancer cell lines, patient-derived xenografts, prostate tumor samples, and CTCs (1, 2, 3). On the other hand, Bernemann and colleagues confirmed that the specified TaqMan-based assays (1, 4) passed the rigorous standardized checklist. These validated assays commonly include PCR primers within exons 4 and 8 in combination with a minor groove binder spanning the junction of exon 4 to 8. These assays detected ARv567es in only three of 45 samples (1) as opposed to the recent report with a high incidence rate of this variant in CTC (78%; ref. 2). This, relatively low detection rate is also observed in our previous study: only 1 of 15 patients was ARv567es positive across the cohort of metastatic prostate cancer (mCRPC; ref. 4). Indeed, in mCRPC, specified exon–exon junction reads by RNA-seq demonstrated low prevalence (4%) of ARv567es, whereas AR-V7 was the most frequent variant (89%; ref. 5). Bernemann and colleagues points to the specific problems of PCR-based ARv567es detection in clinical samples including blood as a potential biomarker. Without confirmatory AR gene sequence and/or protein verification, the incidence and role of ARv567es has yet to be determined.

See the Response, p. 6010