Prognostic Significance of Positive Surgical Margins and Other Implications of Pathology Report

Association of PSMs and Clinical Outcomes

The impact of PSMs on patient clinical outcomes has been studied extensively. PSMs have consistently been associated with elevated risk of PSA recurrence, imparting an approximately 1.5–5-fold increased risk of biochemical recurrence (BCR) (see Table 33.1 ). Additionally, the extent of PSMs affects the risk of recurrence, whereby those with extensive PSMs have a significantly increased risk of BCR in patients compared to those with only focal PSMs. Using a 3 mm cut point, Babaian et al., observed that patients with ≤3 mm of margin positivity had a recurrence rate of 14% compared to 53% in those with >3 mm of margin involvement. Similarly, Lake et al. demonstrated significantly worse 10-year recurrence-free survival when stratified by margin status: 84, 64, 38%, for negative margins, focal PSMs, and extensive PSMs, respectively. Finally, compared to solitary PSMs, Stephenson et al. identified multifocal PSMs as an independent predictor of post-RP disease progression (HR 1.4, 95% CI 1.1–1.8) with significantly worse 7-year PSA progression-free probability of 62% versus 49%. Such observations helped inform the 2009 International Society of Urologic Pathologists (ISUP) consensus statement calling for the amount of tumor at the inked margin to be measured and recorded in millimeters.

An additional PSM characteristic to be documented in the pathology report according to the 2009 ISUP consensus statement is the location of PSMs. Likely due to technical aspects of RP, such as attempts to preserve membranous urethral length and nerve-sparing, the location of PSMs varies anatomically, with observed rates of 8–58% at the apex, 9–40% at the posterolateral prostate, 2–19% at the base, and 4–20% of cases at the bladder neck among cases with PSMs in a review by Fleshner et al. With respect to risk of BCR, all PSM locations do not appear equal. In most studies, apical margins and anterior margins have not been independently associated with increased risk of PSA recurrence, with recurrence rates similar to patients with negative margins. Conversely, posterolateral margins, possibly resulting from attempts at nerve sparing, may be the most significant. In a retrospective study by Eastham et al. examining the outcome of 2442 RPs, isolated posterolateral and posterior PSMs were an independent risk factor (HR 2.8, 95% CI 1.8–4.4 and HR 2.0 (95% CI 1.1–3.6, respectively)) for BCR compared to negative margins whereas other margin sites were not independently associated with BCR in their cohort. Finally, bladder neck margins may be the site with the most controversy. While some studies have reported higher risk of biochemical recurrence others have not. In one recent study, isolated bladder neck PSMs were not associated with BCR when controlling for patient age, PSA, Gleason sore, and pathologic stage, with 12-year biochemical recurrence-free survival (BRFS) similar to men with pT3a and pT3b disease. Similarly, in another study by Buschemeyer et al., isolated bladder neck PSMs imparted a similar risk of recurrence compared to extraprostatic extension (EPE).

Finally, while PSM defines when cancer cells are present at the inked surface, recent studies have evaluated the impact of a close surgical margin challenging the notion that the distance from the tumor to the inked specimen margin does not impact the risk of recurrence. When defined as ≤0.1 mm from the inked margin in >5300 men from the combined series, a close margin was independently associated with a 1.5–2.2-fold significantly increased risk of BCR. Furthermore, the difference in the risk of BCR between close margins and PSMs was insignificant in the study by Izard et al. The proposed mechanism for close margins is that with additional sampling levels, a close margin may actually be positive. Based on these data, although not included in ISUP consensus statements, several institutions now report a close surgical margin in RP specimens as a separate category in addition to positive and negative margins.

While characteristics of PSMs, including extent and location, may influence the risk of BCR, the overall risk of PSMs on more clinically meaningful post-RP outcomes, such as metastatic progression and survival, remains less certain. Largely due to limited follow-up, and the long clinical course of most men with PCa, most studies addressing the impact of PSMs on treatment efficacy have only been able to use BCR. Within these limitations, several recent studies have attempted to address the relationship between PSMs and metastases and death. Specifically, Wright et al. observed a significant 1.7-fold (95% CI 1.3–2.2) increased risk of PCa-specific mortality (PCSM) among men with PSMs versus negative margins among >65,000 men in the Surveillance Epidemiology and End Results cancer registry ( Table 33.1 ). However, on subanalysis, adjusting for adverse pathologic features, these observations remained significant only for men with pT3 and high-grade disease. Similarly, among men with cT3 disease, Mitchell et al. found PSMs at the time of RP to be an independent predictor of PCSM (HR 2.14, 95% CI 1.23–3.72). Finally, while Chalfin et al. confirmed the significant impact of PSMs on PCSM, the impact was small when compared to other factors such as final Gleason score and pathologic stage. Conversely, Boorjian et al. failed to observe an association between PSMs and PCSM or MP, while two studies encompassing >2100 patients combined with >6 and >5 years median follow-up, respectively, demonstrated that PSMs were an independent predictor of metastatic progression (MP) but not PCSM. Thus, while PSMs clearly impact the risk BCR, the overall impact of PSM on the development of clinical metastases and subsequent PCSM remains less clear and is likely in part due to the fact that the natural history of BCR is highly variable, limiting its proxy for disease progression and PCSM. Furthermore, the associated risk and time to event vary widely, dependent to a large extent on the presence or absence of other risk factors. Additionally, attempts to preserve the neurovascular bundles or bladder neck can result in residual benign prostatic tissue as another potential source of measurable PSA irrespective of margin status, confounding the clinical meaning of a detectable PSA following RP, and is in part responsible for the varying definitions of BCR used in the literature. Finally, the protracted natural history of PCa has undoubtedly contributed to the murky waters overlying the relationship between PSMs and metastatic progression and PCSM.

Association Between PSMs and Adjuvant Treatment Outcomes

Existing level 1 data indicate that radiation therapy (RT) delivered to the prostatic fossa immediately following RP may alter the course of high-risk patients with pathologically advanced PCa. In this patient population, the SWOG 8794, European Organization for Research and Treatment of Cancer (EORTC) 20911, and German Cancer Society (ARO96–02) studies demonstrated a 50–60% reduction in the risk of PSA progression among men who received immediate postoperative RT. With longer follow-up (median >10 years), updated results from SWOG 8794, but not EORTC 20911, translated this benefit into a lower risk of metastatic progression (HR 0.7) and improved overall survival in both. While several concerns regarding routine administration of adjuvant radiotherapy following RP persist and reside outside the scope of this chapter (see Chapter 53 for detailed analysis), a subanalysis of EORTC 20911 demonstrated that patients with PSMs on central pathologic review, including patients with poorly differentiated cancers and seminal vesicle invasion, appeared to derive the greatest benefit from adjuvant RT. These data support the notion that, at a minimum, adjuvant RT should be discussed with all men diagnosed with PSMs, in particular perhaps those who have extensive or multifocal PSMs.