Predictive prostate cancer models may identify potential candidates for active surveillance
medwireNews: US researchers have developed predictive models that could help identify men with prostate cancer who are likely to undergo pathologic downgrading after radical prostatectomy and therefore may be suitable candidates for active surveillance.
The cohort study included 2943 men who were classed as grade group (GG)2 intermediate risk at biopsy, of whom 7.6% were downgraded to GG1 after radical prostatectomy, while a further 17.8% had favorable pathology (ie, GG1 or GG2 with <5% Gleason pattern 4).
During a median follow-up of 3 years, the recurrence-free survival (RFS) of men who were downgraded to GG1 and those with favorable pathology was similar to that of the 1325 men grouped as low risk at biopsy, with an 8-year RFS rate of 88%, 89%, and 94%, respectively.
And multivariable analysis adjusted for covariates, such as age at the time of surgery and cancer volume at diagnosis, found no significant difference in RFS between intermediate-risk patients who were downgraded to GG1 and those who were considered low risk at biopsy.
According to researcher Zhuo Su and colleagues from the Johns Hopkins University School of Medicine in Baltimore, Maryland: “Taken together, these findings suggest that prediction of pathological downgrading from biopsy to [radical prostatectomy] helps identify a subset of GG2 [intermediate-risk] patients at diagnosis who may have favorable outcomes more similar to [low-risk] patients and thus potentially justify individual decisions to pursue [active surveillance].”
The reverse pattern – upgrading of risk after radical prostatectomy – was also seen. Among men who were classified as low risk at biopsy, 55.1% had a risk category higher than GG1 and 6.9% had adverse pathologic findings at surgery.
The researchers constructed predictive models using information from multivariable analyses that identified preoperative factors significantly associated with an increased likelihood of pathologic downgrading or favorable pathology at surgery, such as prostate size and cancer volume.
Applying the models showed that 24.7% of men classed as GG2 at biopsy had a greater than 10% probability of downgrading and 37% had a higher than 20% likelihood of having favorable pathology.
Of note, there was no significant difference in RFS between either of these groups of patients and those with low-risk disease at biopsy after adjustment for confounding factors.
Su and colleagues have developed online risk calculators based on these models, which they hope will aid clinical decision making.
They conclude: “Our developed predictive models […] can serve as tools to inform risks of pathological downgrading or upgrading and potentially improve treatment decisions and outcomes among GG2 [intermediate-risk] men who are considering [active surveillance] at diagnosis.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2020 Springer Healthcare part of the Springer Nature group