Prostate Specific Antigen Nadir of 0.1 or Less Is a Predictor of Treatment Success in Men Undergoing Salvage Whole Prostate Gland Cryoablation

Introduction

Cryoablation of the prostate is a viable alternative, to surgical excision, in managing men with local failure of primary radiotherapy for localized prostate cancer. ^{1 –3} Similar to radical prostatectomy, overall survival and disease progression are largely dependent on patient and disease-specific parameters such as Gleason score, timing of relapse, prostate specific antigen (PSA) kinetics, and tumor location. ⁴ Serum PSA levels after salvage cryoablation are a strong predictor of disease progression after treatment. ⁵ In 2016, data from Kovac and colleagues ⁶ demonstrated that men who receive salvage cryoablation have a higher probability of biochemical failure if they fail to reach a nadir PSA below 0.4 ng/mL. This finding provides a helpful cut point for predicting intermediate term progression-free survival; however, these data do not help discern patients with local failure following treatment from patients with distant failure.

Careful patient selection is important in minimizing treatment failure of salvage ablation ⁷ ; however, there are limited data on predictors of local treatment failure among patients who receive salvage cryoablation. Earlier identification of men who develop localized treatment failure may facilitate the identification of men who may benefit from adjunctive therapies. Unfortunately, there are limited data on predictors of local treatment failure after salvage cryoablative treatment. The aim of this study was to determine which perioperative clinical variables predict early treatment failure of men who receive salvage whole gland cryoablation.

Patients and Methods

This is a multi-institutional review of men managed with salvage cryoablation after local or biochemical failure of primary radiotherapy. In total, 925 men were identified through the Cryo-OnLine Database (COLD) registry who received salvage whole gland cryoablation from 1992 to 2002. Men undergoing salvage therapy had positive biopsy following primary radiotherapy. Of these men, 174/925 (18.8%) underwent a per protocol (n = 132) or for cause (n = 42) prostate biopsy following their treatment. The most common primary treatment in these men was external beam radiotherapy (n = 132), followed by brachytherapy (n = 17), and combined external beam/brachytherapy (n = 10). In accordance with practice guidelines, patient with positive seminal vesicle biopsies was not treated with salvage cryoablation. Institutional review board approval was obtained at each study site involved.

Results

Baseline clinical data are presented in Table 1. Fifty-two of 174 (29.9%) of men who underwent biopsy had a posttreatment biopsy demonstrating malignant cancer. Men who had positive biopsy following salvage therapy had significantly higher median nadir PSA (0.7 ng/mL vs 0.1 ng/mL, p < 0.001), shorter median time to prostate biopsy (6.5 months vs 12.0 months, p = 0.001), and shorter median time to biochemical failure (12.1 months vs 25.6 months, p = 0.01). On univariate analysis, the positive biopsy group was more likely to have a presalvage Gleason score >7 (42.6% vs 27.0%) and less likely to have Gleason score <7 (29.8% vs 44.5%); however, these differences were not statistically significant. In addition, PSA before treatment >20 ng/mL was significantly more common among men with positive biopsy (19.2% vs 3.3%, p < 0.001).

A statistical tertile was created to assess the predictive value of different nadir PSA cut points. A tertile was chosen over other statistical quantiles due to the relatively small size of this cohort. A nadir PSA categorical variable was created with cut points at 0.1 and 0.8 ng/mL based on PSA distribution in all men in the cohort. In total, 13/81 (16.1%) men with nadir PSA ≤0.1 ng/mL had positive biopsy, compared to 15/35 (42.9%) and 23/56 (41.1%), with nadir PSA of 0.11–0.8 ng/mL and >0.8 ng/mL, respectively. An adjusted model was constructed (Table 2). Compared to the lowest tertile (PSA nadir ≤0.1 ng/mL), men with PSA in the second tertile (0.11–0.8 ng/mL) and third tertile (>0.8 ng/mL) demonstrated an increased association with a positive biopsy, odds ratio (OR) 4.34 (95% confidence interval [CI] 1.66, 11.4, p = 0.003) and 2.81 (95% CI 1.14, 7.00, p = 0.02), respectively. In addition, a presalvage PSA >20 was associated with an increased risk of positive biopsy (OR 7.65; 95% CI 2.03, 28.9; p = 0.003) following salvage cryotherapy. Similarly, Gleason score 8 or greater was associated with higher odds for positive biopsy, OR 2.26 (95% CI 0.93, 5.47; p = 0.07). Age, time to biopsy, length of time at nadir, prostate volume, and postoperative androgen deprivation therapy were excluded from the final model for p > 0.10. Stratified analysis of 106 men in the cohort who did not receive androgen deprivation demonstrated a positive association with OR 5.31 (95% CI 1.48, 19.1, p = 0.01) and 4.86 (95% CI 1.38, 17.1, p = 0.01), for the second and third tertile, respectively, compared to the first tertile. Among patients who received androgen deprivation, PSA nadir did not correlate with biopsy result at any of the identified cut points (p > 0.10). In addition, an analysis was performed comparing established nadir PSA cut points of significance, such as 0.4 ng/mL, to those identified in our statistical groupings and found the nadir of 0.1 ng/mL to have a more significant correlation with biopsy status after salvage cryoablation.

Discussion

Predicting which patients have local treatment failure after salvage ablative therapy provides opportunity for men to receive additional adjunctive therapies with the intent of impacting disease recurrence, metastasis, and ultimately, disease-specific mortality rate. In our analysis, PSA nadir of 0.1 ng/mL or less is a strong predictor of treatment success. Interestingly, this nadir value has been demonstrated to be associated with PSA recurrence and treatment failure in prior studies. ^{8 –11} More recently, two studies have demonstrated that a PSA nadir above 0.4 ng/mL correlated with biochemical progression. ^6,12 However, a 5-year time period for assessing the clinical end point of recurrence is undoubtedly an admixture of distant and local disease progression in the cohort that was analyzed in their study. We sought to determine if a predictor of local disease failure existed among men undergoing salvage cryoablation assessing the relationships between various clinical patient characteristics and postablation biopsies in men from the COLD registry.

On adjusted models, presalvage PSA and Gleason score independently demonstrated an association with a positive biopsy following treatment. Spiess and colleagues demonstrated that preoperative PSA >5 ng/mL inversely correlated with treatment in success undergoing cryotherapy. ¹³ Similarly, disease aggressiveness before treatment has been shown to be a strong predictor of treatment failure and disease progression in men who receive cryoablative therapy. ⁴ PSA >20 ng/mL was associated with an increased risk of positive biopsy in our cohort. And although not statistically significant, there was an increased odds (OR 2.26; 95% CI 0.93, 5.47) of positive biopsy among men with Gleason score 8 or higher disease at diagnosis. Taken together, it would be reasonable for practitioners to perform biopsy in men with high risk prostate cancer following salvage cryotherapy.

Discerning local failure from distant metastatic disease is difficult after treatment for organ-confined prostate cancer. Finding a predictor of local disease failure is desirable, as it may provide patients who fail salvage cryoablation with an opportunity to receive retreatment with further ablative therapy. This raises a few important questions that our study was not designed to assess. First, there is no guideline recommendation on the timing of PSA measurement following salvage treatment and subsequent prostate biopsy. Our analysis only provides simple descriptive data on the time from treatment to nadir PSA measurement and biopsy, median 3.1 months (IQR 2.8–7.2) and 10.7 months (IQR 6.0–19.6). Based on these observations, we would recommend PSA with or without biopsy around 6 months following salvage cryoablation. Second, we lack information on which adjunctive therapies patients with positive biopsy received and the rates of progression-free survival in these men. Last, it is unclear if patients in this cohort who had positive biopsy also had distant metastatic on bone or chest/abdominopelvic imaging. In fact, our highest nadir PSA tertile (>0.80 ng/mL) was associated with a slightly lower OR of a positive biopsy in comparison to the 0.11–0.8 ng/mL group. This finding suggests that patients with higher nadir PSA values may have nonlocal recurrence of their disease; however, the registry does not have data to fully evaluate the rates of distant metastases in these patients.

There are a few notable limitations that require some discussion. The retrospective nature of our analysis is subject to bias. Specifically, this cohort represents a highly selected population of patients who underwent biopsy both for cause and per protocol. Specifically, only 19% of patients who had salvage cryotherapy underwent posttreatment biopsy. This resulted in a higher proportion of positive biopsy results (29.9%) in the cohort; and if the granularity existed in the COLD registry, it would have been informative to stratify patients by indication for biopsy. It should also be noted that patients with a negative biopsy may be misclassified, and it is difficult to assess the impact of misclassification in this group based on a singular biopsy result. Finally, the study is limited by a smaller sample size and the absence of complete data on important demographic covariates such as ethnicity, type of biopsy (i.e., transperineal vs transrectal), and parameters of salvage treatments. Future analyses assessing treatment success with cryoablative therapy should assess African American ethnicity given the well reported disparities in prostate cancer observed in this group. Nonetheless, this report represents the experience of both academic and community practitioners with expert practitioners; and to our knowledge, this is the first report of a PSA nadir correlating with positive biopsy after disease treatment. This information can help clinicians discern patients at risk for local failure from treatment success following whole gland salvage cryoablation.

Conclusion

Predicting local failure after salvage cryotherapy has the potential to help clinicians identify men who may benefit from additional adjunctive therapy with curative intent. PSA 0.1 ng/mL or less following salvage cryotherapy is predictive of treatment success, and patients with nadir values in this range should not undergo routine biopsy. Consideration should be given to routine biopsy for patients with nadir >0.1 ng/mL after salvage cryoablation. In addition, patients with higher risk disease at diagnosis, such as Gleason score 8 disease and PSA >20 ng/mL, should also be considered for biopsy following treatment given the significant association with positive biopsy observed in our cohort of patients. The authors aim to evaluate this cut point prospectively in men receiving whole gland salvage cryoablative therapy.