Transperineal compared with transrectal biopsy on suspected prostate cancer: a health technology assessment
Health technology assessment
|Published
We have summarized efficacy and safety and performed a health economic evaluation of transperineal biopsy compared with transrectal biopsy for suspected prostate cancer.
Key message
We have summarized efficacy and safety and performed a health economic evaluation of transperineal biopsy compared with transrectal biopsy for suspected prostate cancer. We included four randomized and one non-randomized controlled trial as well as three registry studies. We assessed certainty of the results with the GRADE tool (high, moderate, low, or very low).
We found that transperineal compared with transrectal biopsy may entail:
- Lower risk of infections in general (GRADE: low).
- Lower risk of sepsis (GRADE: moderate).
- Higher risk of hospitalization due to urinary retention (GRADE: low).
- No definite difference in 30-day mortality. The incidence was very low and both fewer and more deaths could occur (GRADE: very low).
- Higher pain intensity in transperineal procedure (GRADE: low).
The latest methods for biopsy, ie targeted biopsy under local anesthesia, were not investigated in the studies on efficacy and safety.
Recent systematic reviews did not indicate that transperineal procedures are inferior to transrectal procedures in the detection of prostate cancer.
According to our economic analysis, higher costs of transperineal procedure can be outweighed by savings resulting from lower costs associated with treating complications.
Summary
Background
On behalf of The Ordering Forum of The National System for Managed Introduction of New Health Technologies within the Specialist Health Service in Norway, the Norwegian Institute of Public Health has conducted a health technology assessment. We have summarized the efficacy and safety and performed a health economic evaluation of transperineal compared with transrectal biopsy in men with suspected prostate cancer or during follow-up after prostate cancer. We have also presented the results from recent systematic reviews that examined diagnostic accuracy of these two biopsy methods.
Traditionally, prostate biopsy performed transrectally has been the most widely used method for suspected prostate cancer. The risk of infections combined with an increased incidence of sepsis and antibiotic resistance in recent years, has led to a proposal to use transperineal biopsy instead. During transperineal biopsy, the needle is inserted through the perineum, thereby avoiding rectal bacterial flora. Traditionally, transperineal procedures have been performed under anesthesia, but the recommendation is to perform the biopsy at an outpatient procedure with local anesthetic. Biopsy sampling can be done systematically in predefined areas or targeted by means of magnetic resonance imaging (MRI) and ultrasound guidance, where the samples are taken of abnormal areas in the prostate. In Norway, the samples are taken in a targeted manner, either cognitively (the MRI image is remembered by the operator) or ultrasound-fused (the ultrasound image and the MRI image are fused). Clinical experts in urology and radiology, as well as patient representatives provided guidance during this project.
Method
We searched for systematic reviews and then for primary studies that covered the inclusion criteria. After inclusion, we assessed methodological quality in the systematic reviews and the risk of bias in the primary studies. The relevant outcomes were infections, sepsis, urinary retention, (re)hospitalizations and pain. We calculated relative risk (RR) with a 95% confidence interval (CI) for dichotomous outcomes. We assessed the most important outcomes with the GRADE tool. Certainty of effect estimates, i.e., whether we have confidence that the effect estimate is close to a true underlying effect, is assessed with the GRADE tool as high, moderate, low, or very low.
We performed a simplified cost impact analysis to compare resource use for transperineal prostate biopsy with that of transrectal biopsy. We also compared costs of treating procedure-related complications of transperineal versus transrectal biopsies.
Results
For efficacy and safety, we included four randomized and one non-randomized study, as well as three registry studies: two large registry studies from England (Berry 2020, n = 73,630) and the United Kingdom (Tamhankar 2020, n = 486,467), and one from Japan. All the included studies used systematic biopsy. Tamhankar 2020 reported total results for the entire ten-year period (2008-2019), but also separately for the final two-year period (2017-2019). Berry 2020 reported results for the period 2014-2017 and adjusted for biopsy time, age, ethnicity, comorbidity and low socio-economic status. These results are also included in Tamhankar 2020. The number of events was very small in the randomized and non-randomized studies. We, therefore, primarily emphasize the results from Tamhankar for the period 2017-2019 when these results are supported by the results from Berry 2020. In most cases, the results were supported by the results from the randomized and non-randomized studies.
We found lower incidence of infections in the transperineal compared to the transrectal group (Tamhankar 2020: urinary tract infections: RR = 0.64 [95% CI: 0.56 to 0.74], general infections: RR = 0.45 [95% CI: 0.39 to 0.51], GRADE: low, table 01). The results for sepsis pointed in the same direction in all studies, but the incidence was higher in the transrectal group compared with the transperineal group (Tamhankar 2020: RR = 0.37 [95% CI: 0.32 to 0.44] Berry 2020: adjusted risk difference aRD = -0.4% [95% CI: -0.6 to -0.2], GRADE: moderate). For hospital readmissions due to urinary retention, the incidence was higher in the transperineal compared with the transrectal group. The results of the randomized and non-randomized trials pointed in the same direction as the results in Berry 2020 and Tamhankar 2020 (Tamhankar 2020: RR = 3.08 [95% KI: 2.61 to 3.63]). The effect was large, but since twice as many samples were collected in the transperineal group as in the transrectal group, and the incidence of urinary retention is assumed to increase with the number of samples the results were not necessarily transferable (GRADE: low). The 30-day mortality rate was 0.07% in the transperineal group and 0.10% in the transrectal group in Berry 2020. In Tamhankar 2020, the percentages were, respectively, 0.05% and 0.07%. The certainty of the evidence was therefore considered to be very low due to few events. Three randomized and one non-randomized study reported pain during and immediately after the procedures. The study with the lowest risk of systematic bias and the highest number of patients found higher pain reported on a visual analog scale in the transperineal compared with the transrectal group (median 4 [IQR 1-6] versus 2 [IQR 0-4]). The results are supported by one of the three other studies (mean [SD] = 8.02 [2.0] versus 5.90 [1.5]), while two reported no difference between the groups. We rated the certainty of the evidence (GRADE) as low.
|
Table 01: Results for adverse events in transperineal compared with transrectal biopsy |
||||
|
Outcome |
Study |
Transperineal percentage |
Transrectal percentage |
RR aRD |
|
Infections |
|
|
|
|
|
· UTI |
Tamhankar 2020 |
0,72% |
1,11% |
0.64 [95% CI: 0.56 to 0.74], |
|
· General |
Tamhankar 2020 |
0.67% |
1.50% |
0.45 [95% CI: 0.39 to 0.51], |
|
Sepsis |
Tamhankar 2020 |
0.42% |
1.12% |
0.37 [95% CI: 0.32 to 0.44]. |
|
|
Berry 2020 |
1.03% |
1.35% |
aRD= -0.4% (95% CI: -0.6 to-0.2) |
|
Urine retention |
Tamhankar 2020 |
0.95% |
0.31% |
3.08 (95% CI: 2.61 to 3.63) |
|
|
Berry 2020 |
1.9% |
1.0% |
aRD = 1.1% (95% CI: 0.7 to 1.4) |
|
30-days mortality |
Berry 2020 |
0.07% |
0.10% |
aRD= -0.03 % (-0.06 to 0.01) |
|
|
Tamhankar 2020 |
0.05% |
0.07% |
|
We included four systematic reviews comparing transperineal and transrectal biopsy and three systematic reviews comparing targeted and systematic biopsy with respect to the detection of prostate cancer. The reviews concluded that the detection rate was comparable for transperineal and transrectal methods, but they called for more high-quality studies to be able to answer the question with greater certainty. Furthermore, three reviews reported that there were indications that targeted biopsy increased the detection rate of clinically significant prostate cancer and reduced overdiagnosis and detection of non-significant prostate cancer compared with systematic biopsy.
According to our economic analysis, higher costs associated with transperineal procedure can be offset by savings associated with lower costs associated with complications. The annual number of cases of procedure-related sepsis can be reduced from 112 cases with transrectal biopsy to 42 cases with transperineal biopsy (for 10,000 biopsies). The measure is cost saving, if it costs a maximum of NOK 1,000 more per procedure than the transrectal alternative. Since there is uncertainty around the cost estimates used in the analyses, the results should be interpreted with caution.
Discussion
The most important knowledge gap that we identified in this HTA is the lack of studies that compared today's most relevant procedures, that is, outpatient targeted biopsy per-formed under local anesthesia (for both transperineal and transrectal method). We therefore lack updated results for the outcomes from such comparisons and especially for transperineal procedure. Outcomes that may be relevant to further investigate are the proportion with urinary retention after transperineal procedure under local anesthesia, pain during the biopsy procedure, and how large a proportion of the population who needs to undergo the procedure in general anesthesia due to pain.
Conclusion
Transperineal biopsy for suspected prostate cancer may reduce the number of infections and sepsis compared to transrectal biopsy. There is possibly a greater risk of urinary retention during the transperineal procedure, but these results are uncertain since the procedures used in the included studies do not fully correspond to the procedures that are relevant in Norway. Based on results from previously published systematic reviews, we did not find any indication that transperineal biopsy is inferior to transrectal biopsy in the detection of prostate cancer. Our economic analysis shows that the higher cost of the transperineal procedure could be offset by savings from lower costs for treating complications associated with this procedure. If the additional costs associated with implementing a transperineal procedure as a standard method for prostate biopsy do not exceed NOK 1,000 per procedure, the intervention can be cost saving for the health service.
