Editorial: Close surgical margins after RP: how to make a complex story even more complex

Surgical margin (SM) status after radical prostatectomy (RP) for clinically localized prostate cancer (PCa) is a measure of surgical quality and retains some prognostic value. Positive SMs (PSMs) have long been considered an adverse oncological outcome because they were repeatedly found to be associated with a higher risk of biochemical recurrence (BCR), and are still among the factors guiding the decision to deliver adjuvant treatments; however, the long‐term impact of PSMs on survival remains uncertain because it is largely affected by other concurrent risk modifiers [1,2,3].

The clinical significance of so‐called close SMs (CSMs), that is, negative SMs (NSMs) with tumour foci approaching, but not involving, the inked cut surface of the RP specimen, is a far less investigated field of research, with contradictory findings in the few available studies (Table 1 [4–12]). Some studies showed a significant association with risk of disease progression (mainly measured with BCR), while others did not.

 

The study by Herforth et al. [12] published in this issue of BJUI further adds to the debate on CSMs, with an analysis of the largest series reported to date. The authors assessed the impact of CSMs vs NSMs vs PSMs after RP on BCR, PCa‐specific and overall survival in ~4 300 men included in the Shared Equal Access Regional Cancer Hospital cohort. CSMs were defined as cancer foci within 1 mm from the inked specimen surface, and were found in 372 patients (9%). The median follow‐up was 6.5 years. On multivariable analysis accounting for several established prognostic factors, CSMs were significantly associated with a higher BCR risk compared with NSMs, but a lower risk compared with PSMs. Notably, SM status alone did not influence PCa‐specific or overall survival. Major limitations to this retrospective analysis were lack of central pathology review and inadequate follow‐up length to assess survival.

The main question yet to answer is whether CSMs entail a biological entity that is distinct from both negative (but not close) SMs and PSMs. Advances in this area cannot be made without taking into consideration the knowledge of PSMs that has accumulated over the past years. We suggest, therefore, that the following principles be adhered to in order to ascertain the true significance of CSMs.

1. 1. Uniform definition
      Some of the available studies used an arbitrary threshold (0.1 or 1 mm) to designate CSMs, but distance between tumour and SMs should be ideally evaluated as a continuous variable before attempting to categorize it.
   2. Accurate pathology examination
      It has been hypothesized that CSMs could be the expression of occult PSMs that are present in different close planes of resection missed by standard sectioning as a result of block sampling bias 11. Encountering CSMs should, then, probably prompt further specimen processing that requires standardization.
   3. Correct prognostic assessment
      It is now accepted that PSMs per se are not sufficient to confer a dismal prognosis, rather it is the concomitant effect of other pathological risk factors (such as stage, tumour volume, Gleason score at SMs, location and extent of PSMs) that determines the aggressive tumour behaviour. The same could apply to CSMs; therefore, their prognostic effect should be investigated by adding ‘interaction terms’ to classic multivariable models that account for a putative synergistic biological effect. It might well be, in fact, that the simultaneous presence of CSMs and extracapsular disease (or higher Gleason score, greater tumour volume, perineural/lymphovascular invasion) results in a final risk of detrimental outcome exceeding the additive combination of the individual risks.
   4. Adequate follow‐up
      At least a decade is required to appropriately test the association of CSMs in patients undergoing RP with endpoints of meaningful interest.

The truth about SMs after RP is still hard to reach, and the issue of CSMs possibly complicates this scenario. While we await further characterization of PCa facilitated by advances in genetic profiling, we recommend that future clinical research in the field does not run into the methodological obstacles of the past.

Gianluca Giannarini, Alessandro Crestani and Claudio Valotto

Urology Unit, Academic Medical Centre ‘Santa Maria della Misericordia’, Udine, Italy