The main objective of the present study was to assess the viability of detecting CTCs in non-metastatic high-risk prostate cancer patients. Our findings revealed a low count and incidence of CTCs. Only 5 out of 65 patients (7.5%) harbored CTCs at diagnosis, that is, at the bottom range of the detection rates of other studies. The few published reports using the CELLSEARCH system in patients with localized prostate cancer showed positivity rates of 5–27% (13–15). These studies were carried out in patients treated predominantly with radical prostatectomy or brachytherapy and included early stages of prostate cancer. In an attempt to improve the CTCs detection rate, we focused the design of the study on including only patients with locally advanced, high-risk disease (13 patients had N1 disease and 30 patients had 2 or more risk factors). To our knowledge, this is the first prospective study performed in this subgroup of patients treated homogeneously with high-dose radiotherapy plus androgen deprivation and with repeated CTCs determinations at predetermined intervals to monitor treatment response.
The many methods used to detect CTCs range from real-time polymerase chain reaction to cell size–based separation or immunomagnetic beads conjugated with anti-EpCAM antibodies (16–19), each of which is subject to intrinsic limitations, including reproducibility. For metastatic prostate cancer, the CELLSEARCH system was shown to provide prognostic information in several large clinical trials (7, 8, 20, 21). Nevertheless, in localized prostate cancer, the CELLSEARCH methodology is now believed to underestimate the actual number of CTCs, perhaps owing to fragmentation of conventional CTCs and the inability to detect the less epithelial CTCs (13). Data from recent studies have shown that EpCAM-based enrichment alone could not detect all CTCs subpopulations (22).
Several attempts have been made to overcome this limitation. Theil et al. (23) used a new system, the CellCollector to isolate in vivo CTCs in patients with different stages of prostate cancer and found more frequent detection of CTCs than with the CELLSEARCH system. Kuske et al. (24) also reported improved detection of CTCs in nonmetastatic prostate cancer patients by combining 3 independent CTCs assays: the CELLSEARCH system, CellCollector, and EPISPOT (an EpCAM-independent enrichment method). Peripheral blood samples were screened for CTCs before radical prostatectomy in 86 high-risk prostate cancer patients and 3 months after radical prostatectomy in 52 patients. The cumulative positivity rate of all 3 CTCs assays was 81.3% (87/107), with 21.5 (23/107) of patients harboring ≥ 5 CTCs per blood sample. The authors hypothesize that the correlation observed with established risk factors and the persistence of CTCs 3 months after surgery could suggest the potential clinical relevance of CTCs as markers of minimal residual disease in prostate cancer.
The second objective of the present study was to analyze whether the variation in the CTCs count at predetermined intervals during the treatment regimen would predict outcome and enable a real-time response. We expected a decline in the numbers of CTCs following treatment. Unpredictably, an increasing and transient CTCs count was detected de novo after androgen deprivation and radiotherapy. The CTCs detection rate was 7.5% at diagnosis, which increased to 12.9% following neoadjuvant androgen deprivation and 18.6% at the end of radiotherapy before decreasing again to 7.5% at 9 months after the end of radiotherapy. When we investigated this changing pattern of CTCs positivization more specifically, we observed a significant association with locally advanced disease (T3-4 stage, p = 0.044; and N1 stage, p = 0.002). Although we do not have a clear explanation for this finding, we believe that it could be due to a passive mechanism associated with the destruction of the tumor. A further analysis with longer follow-up is required to clearly determine the relevance of CTCs positivation during the treatment.
Importantly, other authors have reported similar findings. Stott et al. (25) observed that of 11 patients with preoperative CTCs counts below the cutoff, 4 had transient elevations in CTCs during the follow-up period. Tsumura et al. (12) evaluated whether prostate brachytherapy procedures had a potential risk for hematogenous spillage of prostate cancer cells in 59 patients using the CELLSEARCH system. They detected CTCs from samples immediately after the brachytherapy procedure in 7 patients (intraoperative CTCs detection rates were significantly higher than preoperative ones). However, the authors did not repeat the CTCs analysis during follow-up to evaluate whether those CTCs could actually survive and proliferate at distant sites.
We were unable to show a significant association between positive CTCs status at diagnosis and known clinical and histologic prognostic factors. This lack of correlation has been extensively observed in studies carried out in localized prostate cancer (11, 16, 25). The small sample size and the low CTCs detection rate might limit the weight of these results.
In one of the longest prospective series (152 patients analysed using the CELLSEARCH platform), Meyer et al. (26) did not observe a significant relationship with PSA, Gleason score, or pT stage. The CTCs detection rate in their series was only 11%. Pal et al. (27) used a modified isolation procedure on the CELLSEARCH platform in 35 patients with high-risk, localized prostate cancer. With a CTCs detection rate of 49% prior to surgery, they did not observe any correlation between the presence of CTCs and clinicopathological prognostic features. Conversely, Kuske et al. (24) used the EPISPOT method before radical prostatectomy and found a significant association between CTCs and PSA and clinical T stage. The detection rate with EPISPOT was 58.7%. However, they failed to show any clinical correlation with the other assays (the CELLSEARCH system and the in vivo CellCollector).
Only 1 patient in our series experienced biochemical and distant failure, and 6 patients have since died, none of them from prostate cancer. Again, we did not observe any association between CTCs count and overall survival. Although some authors report a trend towards shorter recurrence times (28), no studies to date have proved the predictive value of CTCs in disease-free or overall survival.
Our study is subject to the limitations inherent to an observational feasibility study, namely, the small sample size and short follow-up. The main strengths are its prospective design with pre-established CTCs determinations at predetermined intervals to monitor treatment response.