Impact of positive surgical margin length and Gleason grade at the margin on oncologic outcomes in patients with nonorgan‐confined prostate cancer

Positive surgical margins (PSM) represent a poor prognostic factor at radical prostatectomy (RP). To investigate the impact of PSM, its length, the focality and the Gleason grade at the PSM, on the oncologic outcomes in nonorgan‐confined RP patients.

margin status, Gleason at the margin and its length carry important information that should be reported for the specimen.

K E Y W O R D S
nonorgan confined, positive surgical margin, prostate cancer, PSM, radical prostatectomy

| INTRODUCTION
[3] Patients with nonorgan confined (pT3) are likely to harbor PSM.Recent studies report rates of PSM ranging from 37% to 61% in pT3 PCa patients. 2,4,5Specifically, the lowest reported rates (37%) of PSM in pT3 patients derived from a robotically treated cohort by Patel and colleagues within a multi-institutional series. 5Conversely, higher rates apply for national registries.Wright et al. 2 reported a PSM rate of 44% for pT3a patients within the Surveillance, Epidemiology, and End Results (SEER) database.Similar rates, compared with the one reported by Wright et al., were also recorded by Lightfoot et al. 4 within a robotically treated cohort for pT3a patients (43%) treated by a single high-volume surgeon.However, for patients with seminal vesicle invasion (pT3b) the authors recorded much higher PSM rates (61%). 4cently, we reported that PSM negatively impacts survival in patients with non-organ confined PCa within the SEER database. 3reover in organ-confined pT2 PCa the Gleason grade at the PSM and its length are poor risk factors for biochemical recurrence (BCR) after RP. 6 Previously, Ploussard and colleagues reported that the PSM length >3 mm and multifocal PSM in non-organ confined prostate cancer are risk factors for BCR after RP. 7 Similarly, Brimo et al. 8 reported within a small cohort of 108 patients with nonorgan that a higher Gleason grade at the margins is associated with BCR.
However, the predictive value of the Gleason grade at the PSM, its length, and the PSM focality in nonorgan confined pT3 disease on long-term oncologic outcomes remains still debatable.
To address this void, we tested the relationship between PSM and BCR, metastasis, and cancer-specific death.Specifically, we examined the relationship of PSM, the Gleason grade at the PSM, its length and focality, and BCR, metastasis and cancer-specific death after RP, within an institutional high-volume center database of contemporary exclusively nonorgan confined PCa RP patients.

| Study population
After Institutional Review Board approval, 5271 patients that harbored pathologic nonorgan confined PCa (pathologic stage T3) were treated with RP between 2010 and 2016 at a tertiary referral institution (Martini-Klinik Prostate Cancer Center, Hamburg-Eppendorf, Germany) were identified.
Exclusion criteria consisted of lymph node invasion (n = 1459) or unknown margin status (n = 105).These selection criteria yielded 3705 patients, which represent the focus of the current study.

| Surgical approach
Surgery was performed either as an open retropubic or robot-assisted laparoscopic approach as previously described. 9,10Neurovascular bundle preservation was performed with the intraoperative frozen section (NEUROSAFE) technique as previously described. 10,11In the case of PSM during the frozen section, a resection of the neurovascular bundle at the corresponding side was performed.

| Surgical margin parameters
Parameters describing the surgical margin status were retrieved from the pathology reports.In cases where a mixed Gleason pattern (i.e., 3 + 4) was recorded for a surgical margin, the predominant Gleason pattern was used for statistical analyses.

| Endpoints
BCR was defined as two consecutive PSA values ≥0.2 ng/ml after surgery.Time to BCR was calculated as the time from RP to BCR or last follow-up.Metastasis was defined as positive imaging following BCR.Imaging procedures consisted of bone scan and/or computed tomography and/or abdominal magnetic resonance imaging and/or 11C-choline positron emission tomography/computed tomography scan.Time to metastasis was also calculated as the time from RP to development of metastasis or last follow-up.Time to cancer-specific death was calculated as the time from RP to death or last follow-up.Cancer-specific death was defined as death attributed to PCa.

| Statistical analyses
Descriptive statistics included frequencies and proportions for categorical variables.Medians and interquartile ranges were reported for continuously coded variables.The χ 2 tested the statistical significance in proportions' differences.The Mann-Whitney U test examined the statistical significance of medians' differences, respectively.Kaplan-Meier analyses graphically depicted BCR-free, metastasis-free (MFS), and cancer-specific survival.Univariable and multivariable Cox regression models were fitted to test the relationship between PSM, Gleason at the PSM, PSM length, PSM focality and BCR, metastasis, and cancer-specific death, respectively.
Patients with adjuvant radiotherapy (n = 362) were excluded from all survival analyses.Specifically, the first set of Cox regression models tested the impact of PSM on BCR, the second set of Cox regression models tested the impact of PSM on metastasis development, the third set the relationship of PSM on cancer-specific death.Subsequently, a fourth and fifth Cox regression model tested the impact of the Gleason patterns at the PSM (Gleason pattern 3 vs.≥4), its length (<4 mm vs. ≥4 mm), and focality (uni-vs.multifocal) on BCR and metastasis in the subgroup of patients with PSM.The adjustment was made for the covariates: age at surgery, year of surgery, preoperative PSA value, Gleason grade groups (GGG 1-2 vs. GGG 3 vs.GGG 4-5), pathologic tumor stage (pT3a vs. pT3b), and pathologic lymph node status (pN0 vs. pNx).R software environment for statistical computing and graphics (version 3.6.2) was used for all statistical analyses.All tests were two-sided with a level of significance set at p < 0.05.

PSM represents a poor prognostic factor at RP in patients with
PCa. [1][2][3] However, the impact of the Gleason pattern at the PSM, PSM length, and the PSM focality in pT3 patients on long-term oncologic outcomes is still under debate.3][14][15][16][17][18] However, the impact of these margin-specific characteristics on hard clinical endpoints, that is, development of metastasis or cancer-specific death is largely unknown.To address this void, we examined the relationship between PSM and BCR, metastasis and cancer-specific death within an institutional high-volume center database of contemporary nonorgan confined PCa RP patients.Our analyses revealed several noteworthy findings.
First, overall we recorded a PSM rate of 27.2% for nonorgan confined PCa.These rates are favorable compared with previous reported PSM rates for pT3 PCa patients, where PSM ranges from 37% to 61%. 2,4,5cond, patients with PSM harbored worse BCR-free survival rates compared with patients with negative surgical margins.BCR-free survival at 96 months after RP was 41.6 versus 57.5% for patients with versus without PSM (p < 0.001).Additionally, PSM represented an independent predictor of BCR (HR 1.53) after multivariable adjustment.In subgroup analysis, focusing only on patients with PSM, multifocal PSMs, a Gleason pattern ≥4 at the margin and a PSM length ≥4 mm were all associated with worse BCR.However, in multivariable analysis, only the PSM length (HR: 1.02, p < 0.05) and a Gleason ≥4 at the margin (HR: 1.60, p < 0.01)    salvage radiotherapy due to BCR after RP. 19 In this study, PSM was no independent predictor for development of metastasis, while the strongest predictor for metastasis following BCR was a higher  20 Our study is not devoid of limitations.First and foremost, it shares limitations of all similar studies that relied on retrospective data.Moreover, our study only provides information from a highvolume PCa referral centers, where all RPs were performed by high-volume surgeons, which may differ from other centers.
Additionally, the site of the PSM may also affect oncologic outcomes after RP, which was unavailable within the current database.Last but not least, our database did not contain information on adjuvant hormone therapy, which may have influenced oncologic outcomes.was an independent predictor for BCR.Next to margin status, Gleason at the margin and its length carry important information that should be reported for the specimen.

F
I G U R E 1 BCR-free survival stratified according to surgical margin status (A), PSM focality (B), PSM Gleason (C), and PSM length (d).BCR, biochemical recurrence; PSM, positive surgical margin [Color figure can be viewed at wileyonlinelibrary.com]

F I G U R E 2
Metastasis-free survival stratified according to surgical margin status (A), PSM focality (B), PSM Gleason (C), and PSM length (D).PSM, positive surgical margin [Color figure can be viewed at wileyonlinelibrary.com] PSM represents an independent predictor for worse oncologic outcomes in non-organ-confined PCa at RP. Gleason ≥4 at the margin was associated with the development of BCR, metastasis, and with cancer-specific death after RP.Moreover, the length of PSM F I G U R E 3 Cancer-specific survival stratified according to surgical margin status (A), PSM focality (B), PSM Gleason (C), and PSM length (D).PSM, positive surgical margin [Color figure can be viewed at wileyonlinelibrary.com]

1
Descriptive characteristics of PCa patients that underwent RP between 2010 and 2016, stratified according to the surgical margins status Abbreviations: GGG, Gleason grade group; IQR, interquartile range; PCa, prostate cancer; PSA, prostatic specific antigen; PSM, positive surgical margin; RP, radical prostatectomy; SD, standard deviation; SM, surgical margin.
Multivariable Cox regression models predicting biochemical recurrence and metastasis in prostate cancer patients with nonorgan confined disease and positive margins at radical prostatectomy 8 A B L E 2 Multivariable Cox regression models predicting biochemical recurrence, metastasis, and cancer-specific death after radical prostatectomy Abbreviations: CI, confidence interval; GGG, Gleason grade group; HR, hazard ratio; PSA, prostatic specific antigen, PSM, positive surgical margin.PREISSER ET AL. | 953 10970045, 2022, 9, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/pros.24341 by Koc University, Wiley Online Library on [11/11/2022].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)onWileyOnlineLibraryforrules of use; OA articles are governed by the applicable Creative Commons License represented independent predictors of BCR.These results corroborate the findings from Ploussard et al., who reported that the PSM length represents a risk factor for BCR in nonorgan confined PCa after RP.7Furthermore, our results underline the report from Brimo et al.8who reported that a higher Gleason grade at the margins is associated with BCR.8Third, patients with PSM more frequently developed metastasis during the follow-up.At 96 months after RP, MFS rates were 82.7 versus 88.6% (p < 0.001) for patients with versus without PSM.However, in multivariable models PSM (HR: 1.15, p = 0.4) failed to reach significance in prediction of metastasis.These results are in line with a previous study within a cohort of patients treated with early T A B L E 3 Abbreviations: CI, confidence interval; GGG, Gleason grade group; HR, hazard ratio; PSM, positive surgical margin.