This week in Urology
Oct 2, 2026
The PSA test for prostate cancer screening: an abridged Cochrane review.
Across six trials and nearly 790,000 men, screening likely prevented about two prostate cancer deaths per thousand, with any all-cause mortality benefit uncertain and conclusions highly sensitive to the threshold chosen.
BJU International · 2026 · PubMed
This week’s highlights
Read the weekly briefing
This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.
Welcome to This Week in Urology. This week we're covering 10 notable papers spanning prostate cancer screening and surgical quality, bladder and kidney cancer treatment, and a cluster of endourology studies on access, positioning and stone-free prediction. Let's dive in.
We start with screening, where two papers approach the same problem from opposite ends. In BJU International, an abridged version of the updated Cochrane review on the prostate-specific antigen test brings together six randomised trials and nearly 790,000 men, with follow-up stretching from three to twenty-three years [1]. Because certainty varied by outcome, the headline estimates lean on a risk-of-bias sensitivity analysis anchored on the European randomised screening trial, where screening likely reduced prostate cancer-specific mortality by about thirteen percent in relative terms. Translated into absolute terms, and that is where the argument actually lives, that is roughly two fewer prostate cancer deaths per thousand men screened, against a baseline of sixteen. For all-cause mortality, the pooled estimate allows for a small benefit but also for essentially no effect, and the authors grade that as low certainty. Adverse events and quality of life showed little to no difference, again with low certainty. The reviewers are explicit that interpretation is highly sensitive to what you accept as a minimal clinically important difference, which is an unusually honest framing of why this debate never quite resolves. They also fold in the ProScreen trial, which paired a kallikrein panel with magnetic resonance imaging: combined screening increased detection of prostate cancer overall, and of both localised and advanced disease, but mortality data are not yet mature, so the central question about that strategy remains open. Against that backdrop, a cohort study in JAMA Internal Medicine asks not whether to screen but whom we are actually screening [2]. Using linked Veterans Affairs and fee-for-service Medicare data covering 1.41 million dually enrolled male veterans aged sixty-five and over, with 8.67 million person-years of observation, Bryant and colleagues found that testing intensity tracked likelihood of benefit more closely inside the Veterans Affairs system. Among men aged eighty and over, prostate-specific antigen testing rates in Veterans Affairs primary care were about thirty percent lower than in fee-for-service Medicare, and biopsy rates about a third lower. Among men with less than five years of predicted life expectancy, Veterans Affairs testing was again lower. Interestingly, the pattern reversed at the other end: among men aged sixty-five to sixty-nine, testing was about ten percent higher in Veterans Affairs care. This is observational and confined to dually enrolled veterans, so it describes system behaviour rather than proving that one system's thresholds are correct, but read alongside the Cochrane estimates it suggests the harm-benefit calculus is being applied unevenly across care settings.
Turning to bladder and kidney cancer, two papers address what to do when standard treatment underperforms. In The Journal of Urology, Gaffney and colleagues report a phase one dose-escalation trial of alternating intravesical gemcitabine and bacillus Calmette-Guérin, which they call GemBCG, in twenty-five patients with bacillus Calmette-Guérin-exposed high-grade non-muscle-invasive bladder cancer [3]. Gemcitabine was escalated from 500 to 2000 milligrams across eight doses, with six doses of 50 milligram TICE bacillus Calmette-Guérin interleaved. There were no treatment-related grade three or higher adverse events and no dose-limiting toxicities, and twenty-four of twenty-five patients had a complete clinical response at six months. Urinary T-cell recruitment chemokines rose and the immunosuppressive cytokine interleukin-6 fell, which is a biologically coherent signal. The caveats are the obvious ones for a phase one study: twenty-five patients, no control arm, a six-month endpoint. The authors note that a randomised phase three trial comparing GemBCG with bacillus Calmette-Guérin retreatment alone is already under way, which is the right place for this question to be settled. On the renal side, BJU International reports a Surveillance, Epidemiology and End Results analysis by Nicolazzini and colleagues of more than sixteen thousand patients with pathological T1b to T2, node-negative, non-metastatic renal cell carcinoma treated with partial or radical nephrectomy [4]. About eighteen percent waited more than three months from diagnosis to surgery. After propensity score matching, five-year cancer-specific mortality was 7.6 percent in the delayed group versus 5.9 percent, and in competing-risks regression delay independently predicted roughly a thirty percent higher cancer-specific mortality. There was a dose-response gradient, with the effect larger beyond six months, and the association held across tumour grades but was not evident in non-clear cell disease. This is registry data with all the residual confounding that implies — sicker patients wait longer — but the gradient and the histology-specific pattern lend it some credibility as evidence that delay in clear cell disease is not biologically neutral.
Two papers from the same high-volume robotic prostatectomy programme tackle surgical quality from complementary angles. In BJU International, Bilé-Silva and colleagues report on 1385 men undergoing robot-assisted radical prostatectomy within a patient-specific surgical planning pathway, stratified by the number of STAMPEDE-defined high-risk factors [5]. Among the 262 men with two or more high-risk features, some form of nerve-sparing was performed in just over half. Positive surgical margin rates and twelve-month biochemical recurrence rates were statistically indistinguishable between nerve-sparing and non-nerve-sparing surgery, and pad-free continence rates at three and twelve months were also similar — though men in the nerve-sparing group used significantly fewer pads at twelve months. This is retrospective, single-centre, and crucially the nerve-sparing patients were selected for it, so it establishes feasibility within a structured planning pathway rather than demonstrating safety for high-risk disease generally; the authors call for prospective evaluation. The companion question — where margins actually come from — is addressed in World Journal of Urology, where Day and colleagues performed a pathway-wide root-cause analysis of 100 consecutive patients with positive margins, tracing 112 margins back to 125 contributing sources across diagnosis and staging, surgical planning, and surgical execution [6]. Just under eighty percent of sources were attributed to surgical execution, with nerve-sparing implicated in about half of all positive margins and apical dissection in roughly thirty percent; diagnostic and planning factors together accounted for about a fifth. That is a useful tension to hold alongside the first paper: nerve-sparing appears oncologically acceptable in aggregate in selected high-risk men, yet it remains the single most frequently implicated step when margins do occur. The authors frame this as the case for intraoperative margin assessment and as a quality-improvement framework, not as an argument against nerve-sparing.
Finally, four endourology papers. The most substantial is a two-centre randomised trial in European Urology Focus, where Shen and colleagues compared dual contrast-enhanced ultrasound guidance against fluoroscopy for renal access in 150 patients undergoing percutaneous nephrolithotomy [7]. First-attempt puncture success was ninety-two percent with the ultrasound approach versus eighty percent with fluoroscopy, roughly tripling the odds of success on the first pass. Median puncture time fell from two minutes to under one, tract establishment time was halved, median hospital stay was two days versus five, and haemoglobin loss was lower. Complication rates did not differ significantly. The trial was open-label, which the authors concede limits robustness, and the dramatic hospital stay difference invites questions about local practice patterns — but as a radiation-free alternative this now has randomised support and, in the authors' words, warrants further evaluation. Complementing that, a Young Academic Urologists and European Association of Urology endourology section meta-analysis in the Journal of Endourology pooled twenty-two studies and nearly nine thousand procedures to find a primary retrograde ureteral access failure rate of about eleven percent, with heterogeneity that was frankly substantial [8]. Younger age, male sex and proximal stone location were associated with higher failure, while thinner ureteroscopes and prior ipsilateral stone surgery were protective. The evidence is retrospective with signs of publication bias, so the authors position these as counselling and decision-making aids rather than firm selection criteria. In World Journal of Urology, Wang and colleagues pooled eight comparative studies and 998 patients and found that intraoperative positioning strategies during retrograde intrarenal surgery raised stone-free rates by about twenty percent in relative terms compared with standard lithotomy, consistently in lower pole and mixed intrarenal stones, with no significant difference in postoperative fever [9]. Operative time and length of stay data were too heterogeneous to interpret, and there was no intrarenal pressure monitoring. And rounding out the group, Durak and colleagues in the same journal developed a preoperative score predicting clinically significant residual stone after mini-percutaneous nephrolithotomy in 217 patients, where body mass index, stone number, volume, density and lower pole location were independent predictors [10]. Discrimination was high, with an area under the curve of about 0.96 and an optimism-corrected value of 0.95, numerically outperforming the S.T.O.N.E. and Guy's scores — but this is single-centre with internal validation only, and the authors are clear that external multicentre validation is required before broader use.
If you only have time for one paper this week, make it the abridged Cochrane review of prostate-specific antigen screening in BJU International [1]. It is the most rigorous current statement of what screening does and does not deliver, and its central point — that the whole conclusion pivots on what you accept as a clinically meaningful difference — is the honest frame for every shared-decision conversation about screening.
Here is what this week's evidence adds up to in Urology. First, screening likely reduces prostate cancer-specific mortality by about two deaths per thousand men, with any all-cause mortality benefit uncertain, and the newer kallikrein-plus-magnetic resonance imaging strategy has shown more detection but no mortality data yet [1]. Second, large observational data suggest testing intensity is better matched to likely benefit in Veterans Affairs primary care than in fee-for-service Medicare, particularly in men over eighty and those with short life expectancy [2]. Third, adding gemcitabine to bacillus Calmette-Guérin retreatment looked safe with a striking six-month response rate in a twenty-five patient phase one trial, and the randomised answer is pending [3]. Fourth, registry data link surgical delay beyond three months to higher cancer-specific mortality in clear cell renal cell carcinoma, with a dose-response gradient but unmeasured confounding [4]. And fifth, across the endourology papers, ultrasound-guided access now has randomised support for first-attempt success without radiation [7], while the access failure risk factors, positioning strategies and residual stone score all rest on lower-quality or single-centre evidence that is not yet guideline-grade [8][9][10].
That's your roundup for This Week in Urology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.
If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if it saves you time, a quick rating in your podcast app helps other physicians find the show.
This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.
Get Urology every week
By email, Spotify, Apple or any podcast app.
Spot something worth flagging?