The surgical learning curve and post-RALP outcomes over time

Two new and quite separate papers offer rather different perspectives on the value of robot-assisted laparoscopic prostatectomy (RALP) as a first-line treatment for the management of localized prostate cancer.

One of these papers, by Thompson et al., offers a detailed analysis of the experience of a single, Australian, high-volume prostate cancer surgeon. The surgeon in question conducted  1,552 consecutive RALPs on patients between 2006 and 2012. Prior to that, he already had accumulated experience of conduction about 3,000 open radical prostatectomies (ORPs). The full text of this paper is available on line.

Although this is a single-surgeon study, it is clear that the research team used a very careful, prospective, data collection process over time. a wide variety of demographic and clinicopathologic data were collected; quality of life questionnaires were administered to as many patients as possible at baseline, and at 1.5, 3, 6, 12, and 24 months post-surgery.

The basic results of this study are as follows:

  • Data from 1,511 patients were included in the positive surgical margin (PSM) analysis.
  • Data from 609 patients were included in the quality of life (QoL) analysis.
  • Post-RALP sexual function scores
    • Surpassed post-ORP sexual function scores after just 99 RALPs
    • Increased to a mean difference of 11.0 points with the 861st patient
    • Plateaued at between 600 and 700 RALPs.
  • Post-RALP early urinary incontinence scores
    • Surpassed post-ORP early urinary incontinence scores after 182 RALPs
    • Increased to a mean difference of 8.4 points
    • Plateaued at between 700 and 800 RALPs.
  • With respect to risk for PSMs
    • The risk for a pT2 PSM was initially higher after a RALP than after an ORP.
    • The risk for a pT2 PSM was lower after a RALP than after an ORP after 108 RALPs.
    • The risk for a pT2 PSM was 55 percent lower after a RALP than after an ORP after 866 RALPs (odds ratio [OR] = 0.45).
    • The risk for a pT3/4 PSM was also initially higher after a RALP than after an ORP.
    • The risk for a pT3/4 PSM after a RALP decreased over time, plateauing at around 200 to 300 RALPs
    • The risk for a pT3/4 PSM was still 15 percent higher after a RALP than after an ORP after 866 RALPs (OR = 1.15).

Thompson et al. draw a series of significant conclusions from these data:

  • There is a long learning curve associated with the use of robot-assisted technology in the conduct of RALPs (just as there is for every other type of complex surgery)
  • Even experienced surgeons should expect to have inferior outcomes initially.
  • Over time, experience should be associated with progressively superior sexual, early urinary, and pT2 PSM outcomes and similar pT3 PSM and late urinary outcomes.
  • Learning do do RALPs well over time was worthwhile for this high-volume surgeon, but …
  • This learning curve may not be justifiable for late-career/low-volume surgeons.

The other paper, by Hu et al., offers a retrospective observational analysis of data from 5,556 RALPs and 7,878 ORPs carried out between 2004 and 2009, based on data from the Surveillance Epidemiology and End Results (SEER)–Medicare database. In this case, only the abstract of the paper is available on line.

Studies of this type are statistically complex, and always need to be interpreted with caution because they are based on so-called “administrative” (as opposed to clinical databases. For example, Hu et al. had no data related to PSA levels post-surgery by which they could assess risk of biochemical recurrence. However, using a statistical system known as “propensity-based analysis” to minimize treatment selection biases, here are the core findings of Hu and his colleagues:

  • RALP was associated with less risk for PSMs than ORP (13.6 vs 18.3 percent; OR = 0.70).
  • There were fewer PSMs among the RALP-treated as opposed to the ORP-treated patients who had
    • Intermediate-risk disease (15.0 vs 21.0 percent; OR: 0.66)
    • High-risk disease (15.1 vs 20.6 percent; OR = 0.70)
  • RARP was associated with less use of additional forms of cancer therapy than ORP
    • Within 6 months of surgery (4.5 vs 6.2 percent; OR =0.75)
    • Within 12 months of surgery (OR = 0.73)
    • Within 24 months of surgery (OR = 0.67); 95% CI, 0.57–0.78) of surgery.

The “take home message” from this paper is that, although RALPs are more expensive than ORPs, RALPs are “associated with fewer positive surgical margins and less use of additional cancer therapy within 2 yr postoperatively.”

If one considers these two papers in concert, it seems to The “New” Prostate Cancer InfoLink that the overall message here is that when skilled, high-volume surgeons have learned to do RALPs well (i.e., after at least 100 and probably more like 200 patients), then their general level of outcomes start to be better than those achievable by skilled, high-volume surgeons doing ORPs. However, there are always going to be exceptions to a general rule like this.

From a patient perspective, however, if one wants to have surgery as first-line treatment for prostate cancer, what is important are the data specific to the individual surgeon you are talking to. Not every surgeon is going to be able to provide patients with the type of detailed data reported by Thompson et al. … but it would be inappropriate to think that every surgeon who has done 200 or even 1,000 RALPs is necessarily doing them well. Volume on its own is not enough. What matters is both the volume and the basic skill levels of the surgeon, combined with his or her determination to keep trying to do this operation better over time. As noted by Thompson et al., the surgeon associated with this study freely admits that it took him more like 600+ patients to reach a consistent plateau of outcome quality with optimal levels of post-surgical continence and negative surgical margins, and he started out with some 3,000 ORPs under his belt!

It becomes more and more evident over time (if it wasn’t to begin with) that radical prostatectomy is a very highly specialized surgical procedure, best conducted by highly skilled, subspecialized surgeons, working at high-volume centers with close attention to their outcomes data over time. This does not by any means imply that there aren’t excellent surgeons working out in community hospitals, but it does mean that we have far less information about their skills and outcome levels.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: