Badar M. Mian, MD

Intravesical instillation of BCG remains the most frequently utilized therapy against high-grade non–muscle-invasive urothelial bladder cancer (NMIBC). Treating urologists are quite familiar with the identification and management of early urinary and systemic adverse effects of BCG instillation. However, the spread of BCG infection to other organs is less well appreciated, likely due to the low incidence of such complications. Further, the published literature mostly consists of single-case reports. To increase awareness of these complications, Cabas et al performed a systematic review of the published literature demonstrating the frequency and timing of potentially serious post-BCG infections involving genitourinary (GU) and non-GU sites.

To identify all case reports of post-BCG infections, the investigators systematically reviewed 401 publications through Medline and other searches. Of these, 271 case reports, including 307 patients, met the inclusion criteria for the review. From these reports, the authors compiled data including demographics, tumor grade/stage, number of intravesical instillations, number of organs with BCG infection, time from last BCG instillation to signs of infection, type of imaging utilized, microbiological confirmation, type and duration of surgical and nonsurgical treatment, and patient outcomes.

Of the 307 cases of post-BCG infection, only 4 (1.3%) occurred in female patients. The average number of intravesical BCG instillations prior to developing symptoms of infection was 8 plus or minus 5.3 instillations. A single organ/site was infected in 72.0%, 2 organs were affected 20.2%, and 3 or more organs were affected 7.8% of the cases. Of all the cases identified by the authors, including those with multiple organs involved, BCG-related infection of GU organs occurred in 23.7% and non-GU organs were involved in 76.2% of cases. Most frequently involved sites included non-GU organs such as lungs, vascular system, and liver, whereas the various GU organs (testis, penis, prostate, and kidneys) were involved equally and infrequently.

The median time from last BCG instillation to clinical signs of infection varied and depended on the organs involved. The granulomatous infection of the liver and lungs (such as miliary pulmonary infection) occurred earlier, at a median time of 1 week from the last instillation. In contrast, vascular lesions (endograft graft infection or mycotic aneurysm), bone/joint involvement (prosthetics infection or spondylodiscitis), and muscular involvement (psoas abscesses) occurred late, at a median of 52, 68, and 93 weeks, respectively. Amongst the GU organs, penile lesions (erythema, ulcer, granulomatous lesions) appeared at a median of 1 week, prostatitis or abscess at 3 weeks, renal granulomatous disease at a median of 8 weeks, and epididymo-orchitis at a median of 56 weeks.

Overall, microbiological diagnosis was made in 55.6% of cases through acid-fast stain, culture, or polymerase chain reaction test. The rate of isolation and mycobacteria identification at the affected sites also appears to be related to the organs involved. Microbiological detection was noted in 28.6% of penile lesions, 35.2% of lung lesions, and nearly 60% of prostate/epididymis/testis lesions. The highest rates of microbiological detection of the mycobacteria were noted in vascular (83.6%), bone/joint (91.7%), and muscle lesions (100%). In evaluating concomitant infections at various sites, the investigators observed a strong association (

 < .001) between lungs and liver complications, bone marrow and liver complications, bones/joints and muscles, and lungs and bone marrow complications. No significant associations between concomitant infections and various GU organ sites were identified.

The reported duration of antimycobacterial therapy ranged from a median of 6 weeks for penile or testicular lesions, 8 weeks for lungs or liver lesions, and 12 weeks for bone/joint infections. Surgical intervention was most frequently required for cases of infections involving vascular (83%), bone/joint (85.4%), and muscular systems (90.5%), whereas surgical intervention was relatively infrequent for lung (6.8%) and liver (8.9%) infections.

Of the GU organs involved with post-BCG infection, a majority (55%-90%) of the penile, prostate, and kidney infections manifested within 1 month of the last BCG instillation, whereas testicular cases appeared after 12 months or longer. For the non-GU organs, there appears to be a biphasic distribution of the time to post-BCG infection. A majority (50%-75%) of the lung, liver, and bone marrow infections became apparent within 1 month. More than 50% of the infections involving vascular, muscular, and bone/joint systems appeared 12-24 months after the last BCG instillation.

An interesting observation from this compilation of 307 case reports is that only 1.3% of post-BCG systemic infections occurred in female patients. One may speculate that the male anatomy (bulbar urethra, prostate) is more prone to catheter-related trauma during instillation, followed by increased exposure of traumatized area to the BCG when voiding. Both of these would result in potential intravasation of BCG resulting in systemic spread of BCG. Thus, strict adherence to the protocol for canceling the BCG in the presence of infection, and aborting the instillation in the setting of trauma and/or any degree of gross hematuria, is required. Although systemic infections involving non-GU organs are uncommon, these infections are associated with significant morbidity, and mortality. Of the 65 cases of vascular involvement, 10 (15.4%) died of BCG-related complications. The mortality rate for the post-BCG infections involving lungs, liver, and muscles was reported to be 4.1%, 3.6%, and 4.8%, respectively.

This study provides clinically useful data on the frequency and timing of BCG-related infection of various organ systems. Urologists play a pivotal, long-term role in the diagnosis and treatment of NMIBC, and essentially serve as the primary oncologists for patients with NMIBC. It is essential for us to be vigilant about all the complications, whether local or systemic, associated with intravesical BCG. A high index of suspicion is required for early identification of these infrequent, but potentially lethal, post-BCG systemic infections.

1. Cabas P, Rizzo M, Giuffrè M, et al. BCG infection (BCGitis) following intravesical instillation for bladder cancer and time interval between treatment and presentation: a systematic review. 

2021;39(2):85-92. doi:10.1016/j.urolonc.2020.11.037

Articles

"It is essential for us to be vigilant about all the complications, whether local or systemic, associated with intravesical BCG," writes Badar M. Mian, MD.

Urologists must remain vigilant regarding post-BCG infection

offers perspective on noteworthy research in the peer-reviewed literature. Dr. Mian is professor of surgery in the division of urology at Albany Medical College, New York.

Despite a paucity of high-quality evidence, extended pelvic lymph node dissection (EPLND) at the time of radical prostatectomy (RP) has been being recommended with increasing frequency over the last few years. Several retrospective studies have reported increased lymph node yield as well as increased detection of positive lymph nodes. However, the true extent of any therapeutic benefits of EPLND remains debatable. In a recently reported randomized study comparing limited pelvic lymphadenectomy (LPLND) and EPLND, Lestingi et al demonstrated that although EPLND may improve pathological staging, the oncological benefits remain unclear.

The study enrolled 300 patients from May 2012 to December 2016, and randomly allocated 150 men to LPLND and 150 men to EPLND. The primary end point of interest was 5-year biochemical recurrence-free survival (BRFS). Secondary end points included metastasis-free survival (MFS), cancer-specific survival, and histopathological findings. The study was designed to detect a 15% improvement in 5-year BRFS in the EPLND group, assuming a 5-year BRFS rate of 65% for the LPLND group.

For the entire cohort, median age was 63 years and median prostate-specific antigen level was 10.5. Gleason grade group distribution showed 36% of patients were grade group (GG) 1, 40% GG 2, and 23% GG 3 to 5. Nearly 62% were classified as D’Amico intermediate-risk prostate cancer and 38% as high-risk prostate cancer. For the bilateral lymph node dissection, LPLND was confined to the obturator zone only while the EPLND included the obturator, external iliac, internal iliac, common iliac, and presacral regions.

After RP and pelvic lymphadenectomy, pathologic stage T3a or higher was noted in more than 60% of patients, and positive surgical margins were noted in 40%. The median number of nodes removed with LPLND and EPLND was 3 and 17, respectively (

<.001). Lymph node metastases were noted in 5 (3.4%) men in the LPLND group and in 25 (17%) men in the EPLND group (

<.001). In subgroup analysis of men with D’Amico intermediate-risk and high-risk prostate cancer, a higher lymph node metastases rate was noted with EPLND (11% and 26%, respectively) than with LPLND (0% and 11%). Of the 25 patients with lymph node metastases in the EPLND group, the most frequent sites included internal iliac (65%), external iliac (39%), and obturator (35%) lymph nodes.

The median follow-up for patients without biochemical recurrence was 53.9 months. No difference in BRFS was noted between EPLND and LPLND groups. The median BRFS was 61.4 months in the LPLND group, and it was not reached in the EPLND group (HR, 0.91; 95% CI, 0.63-1.32; 

=0.6). Further, the median MFS was not reached in either group, and there were no differences between the 2 groups in terms of distant metastases or death. Salvage radiation therapy utilization was similar in both groups (32% in LPLND, 25% in EPLND). Androgen deprivation therapy was required in 18% and 14% of men in the LPLND group and the EPLND group, respectively.

In the subgroup of men with biopsy GG 3 to 5 undergoing EPLND, the BRFS was superior to men undergoing LPLND using both the intention-to-treat (HR, 0.33) and per-protocol analyses (HR, 0.48). Using Kaplan-Meier estimates, intention-to-treat analysis showed median BRFS at 12.34 months in the LPLND group and not reached in the EPLND group (HR, 0.48; 

=0.02). However, in the per-protocol analysis, the median BRFS in the LPLND and EPLND groups was not statistically significant (19.9 months vs 32.7 months, respectively; HR, 0.79; 

One of the limitations of the study is the somewhat short follow-up to assess survival outcomes; however, this follow-up may be acceptable for the early oncological outcome such as biochemical recurrence. Further, the complications associated with PLND were not reported and will likely be revealed in a follow-up manuscript. Although findings suggest that a subset of patients, such as those with Gleason GG 3 to 5, may benefit from EPLND, the number of subjects in each group was too small (LPLND, 38 men; EPLND, 31 men) to draw firm conclusions. The oncological benefit of EPLND in men with high-grade prostate cancer would require validation in a separate larger study.

Previous publications (typically observational or case-control studies) have variably reported some oncological benefit associated with EPLND but also increased risk of complications such as lymphocele formation, bleeding and increased operative time. The current study may represent the only study providing a high level of evidence about the oncologic benefits of EPLND at the time of RP. In this randomized trial, EPLND significantly improved lymph node staging but failed to demonstrate any overall benefit compared to LPLND in terms of biochemical recurrence or metastases.

While we await results of other trials, especially those including a larger number of men with biopsy GG 3 to 5, or high-risk prostate cancer, it appears that EPLND is unnecessary in most patients undergoing RP, with the possible exception of men with high-risk disease.

1. Lestingi JFP, Guglielmetti GB, Trinh Q-D, et al. Extended versus limited pelvic lymph node dissection during radical prostatectomy for intermediate- and high-risk prostate cancer: early oncological outcomes from a randomized phase 3 trial. 

Published online December 5, 2020. doi:10.1016/j.eururo.2020.11.040

The technique fails to demonstrate benefits for prostate cancer recurrence and metastases.

Extended pelvic lymph node dissection during RP does not appear beneficial

Active surveillance (AS) strategies for prostate cancer with low-risk features, such Grade Group 1 (GG1), are increasingly being utilized to avoid treatment-related complications. Due to the long natural history of prostate cancer and low risk of progression, long-term risk of metastasis or death is not well defined, resulting in some uncertainty about patient selection for AS. A recent study by Maggi et al noted that although reclassification of the initial biopsy Grade Group or switch from surveillance to active treatment occurred frequently, less than 1% of men died of prostate cancer.

 They analyzed the data from a large, prospectively maintained AS cohort to identify the variables associated with disease progression and metastases while on AS.

The investigators identified 1450 men on AS between 1990 and 2018 who met the inclusion criteria of low- or intermediate-risk prostate cancer. AS eligibility criteria included organ-confined disease with biopsy GG1, 33% or less positive biopsy cores, and PSA density (PSAD) lower than 0.15 ng/mL/cm3. However, few selected patients with GG2 and/or higher PSAD were also enrolled into the AS program. 1303 patients (90%) were diagnosed with initial biopsy GG1 and 147 (10%) with GG2. Median PSA at initial biopsy was 5.4 ng/mL, with median PSAD of 0.13 ng/mL/cm3. Median age was 62 years and median follow-up was 77 months. Additional testing such as Genomic Prostate Score (GPS) was performed in 34% of men (mean GPS 25), whereas multiparametric magnetic resonance imaging (mpMRI) was performed in 46% of men, of whom 24% had lesions classified as Prostate Imaging Reporting and Data System (PI-RADS) 4-5.

The primary outcome of interest was emergence of metastatic prostate cancer noted on imaging or pathology of delayed prostatectomy. Secondary outcomes included upgrading on surveillance biopsy to GG2 or greater, receipt of active treatment, and prostate cancer-specific and overall survival. For this analysis, patients were divided into 3 subgroups based on the initial biopsy GG and PSA. Of these, GG1 with PSAD lower than 0.15 was noted in 57%, GG1 with PSAD of 0.15 or higher in 32%, and GG2 in 10%.

At 7 years, biopsy GG upgrade-free survival was 40%, and median time to biopsy GG upgrade was 25 months. The 7-year upgrade-free survival was higher in patients with initial GG1 with PSAD lower than 0.15 (47%) compared with those with GG1 with PSAD of 0.15 or higher (28%) or initial GG2 (44%) (log rank 

 < .01). The 7-year treatment-free survival was 59%, with a median time to receiving active treatment of 32 months. The 7-year treatment-free survival was higher in patients with initial GG1 with PSAD lower than 0.15 (69%) compared with those with GG1 with PSAD of 0.15 or higher (46%) or initial GG2 (42%) (log rank 

At 7 years, metastasis-free survival was 99%, and median time to metastases was 62 months. Of the 15 patients who developed metastases, 2 were lost to follow-up, 9 had lymph node metastasis, 2 had lymph node and bone metastasis, 1 had bone metastasis, and 1 had bone and visceral organ metastasis. Initial biopsy GG2 disease was associated with lower 7-year metastasis-free survival (96%) than those with GG1 with PSAD lower than 0.15 (99 %) and with GG1 with PSAD of 0.15 or higher (100%) (log rank 

 < .01). On multivariable analysis, risk factors for increased risk of metastases included initial GG2, PSA velocity, and PI-RADS 4-5 lesion. However, age, PSAD, and GPS did not increase the risk of metastases. Of note, in those who did not develop metastases, the median follow-up was quite sufficient at 77 months.

Overall, 64 of 1450 patients died at a median of 84 months. The 7-year overall survival was 97% for the entire cohort. Overall survival rates differed by sub-groups: GG1 with PSAD lower than 0.15 (98%) compared to those with GG1 with PSAD of 0.15 or higher (96%) or initial GG2 (87%). Only 4 of 64 patients in this AS program died of advanced prostate cancer. The 7-year prostate cancer-specific survival was greater than 99%.

This report is based on one of the few large AS cohorts with long-term follow-up. Although 60% of patients experienced upgrading on subsequent biopsies, only 41% underwent deferred treatment, suggesting that upgrading alone may not necessitate immediate treatment. The goal of AS protocols is to avoid or delay treatment for those who may not benefit from it, and still maintain the curative intent through early capture of any disease progression. The success of AS protocol is evident from the data showing that 1% of men with low-risk prostate cancer developed metastases, and only 0.27% died of prostate cancer.

Further details about other prognostic variables in patients with initial GG2, such as age, focality, or percentage of Gleason pattern 4, would have been instructive in identifying the optimal candidates for AS. Additionally, this AS protocol included variable diagnostic testing and surveillance protocol during the study period. Should patients with GG2 on AS be managed with a more strict or intensive surveillance program than those with GG1? Although most of the patients in this study were Caucasian (92%) and the general applicability may be questioned, recent studies have highlighted the safety of AS in African American men with low-risk prostate cancer.

The boundaries of AS inclusion criteria are being expanded to include selected men with GG2 prostate cancer. Incorporating mpMRI, and perhaps genomic tests, will further facilitate the selection of appropriate candidates for AS, thus increasing its acceptability by urologists and patients. Despite some heterogeneity in the type and frequency of testing during the surveillance period, these results provide reassurance and increased confidence in the safe utilization of AS protocols for men with low-risk GG1, and selected GG2, prostate cancer.

1. Maggi M, Cowan JE, Fasulo V, et al. The long-term risks of metastases in men on active surveillance for early stage prostate cancer. 

. 2020;204(6):1222-1228. doi:10.1097/JU.0000000000001313

"Despite some heterogeneity in the type and frequency of testing during the surveillance period, these results provide reassurance and increased confidence in the safe utilization of AS protocols for men with low-risk GG1, and selected GG2, prostate cancer," Mian writes.

Active surveillance data in prostate cancer show frequent upgrading, rare mortality

Despite previous randomized controlled studies demonstrating a benefit from post-prostatectomy adjuvant radiation therapy (RT), urologists have been slow to adopt this approach. This may be related to inconsistencies in the reported outcomes as well as the timing of salvage RT, which was thought to have been initiated much later than the contemporary clinical practice. Consequently, 3 randomized controlled trials (RADICALS-RT, GETUG-AFU 17, and RAVES) were initiated nearly a decade ago to evaluate the oncologic outcomes after adjuvant RT or early-salvage RT in post-prostatectomy patients with high-risk and/or locally advanced pathology. Recently, a combined analysis of these 3 trials was reported, which did not demonstrate any advantage to adjuvant RT when compared to early-salvage RT; that is, following rise in prostate-specific antigen (PSA) level after prostatectomy.

Vale and colleagues prospectively planned a systematic review and meta-analysis of the combined data from the 3 trials before the final results were available to assess whether adjuvant RT was superior to early-salvage RT.

 Individually, the 3 trials were powered for different primary outcomes such as time free of metastases (RADICALS-RT), event-free survival (GETUG-AFU 176), and biochemical progression (RAVES). Inclusion criteria for the trials varied slightly but included the following: postoperative PSA less than 0.2 ng/mL, and one or more high-risk features including stage pT 3 or 4, Gleason score 7–10, or positive surgical margins.

After randomization, adjuvant RT was to be initiated within 6 months after surgery while early-salvage RT for biochemical recurrence was to be initiated at PSA level of 0.2 ng/mL (0.1 ng/mL for RADICALS-RT). The RT dose was quite similar in all 3 trials; that is, 64 to 66 Gy in 32 or 33 fractions. The concurrent use of hormone therapy was variable across the 3 trials. For this meta-analysis, the agreed-upon primary outcome measure was event-free survival. An event after randomization was defined as the first evidence of any 1 of the following: biochemical progression (PSA ≥ 0.4 ng/mL), clinical or radiological progression, initiation of additional treatment, and PSA level of at least 2.0 ng/mL at any time after randomization.

In the 3 trials combined, 2153 patients were recruited between November 2007 and December 2016. Of these, 1075 patients were randomly assigned to receive adjuvant RT and 1078 were assigned to early-salvage RT. Median age was 64 to 65 years. A majority of patients had either stage pT3a/b disease (79.8%), positive surgical margins (70.9%), and extracapsular extension (76.9%). The median follow-up ranged from 60 to 78 months.

There were 270 (12.5%) events recorded out of the 2153 randomized patients. The pooled event-free survival analysis yielded an overall hazard ratio of 0.95 (95% CI, 0.75-1.21; 

= 0.70) which translates into only a 1% absolute difference between adjuvant RT and early-salvage RT groups. Further, there was no advantage to adjuvant RT on event-free survival when analyzed based on pre-specified subgroups including pre-surgical PSA, Gleason score, seminal vesicle involvement, surgical margins, or CAPRA-S risk group. In the group assigned to early-salvage RT, only 421 (39.1%) had required RT at the time of this analysis. A majority of the men remained free of biochemical progression at 5 years (87% in RAVES, 88% in RADICALS-RT, and 94% in GETUG-AFU 17).

It seems clear from this meta-analysis that adjuvant RT for men with high-risk, localized, or locally advanced prostate cancer after prostatectomy does not improve event-free survival. These results are in contrast to the previous trials (from EORTC and SWOG), which had demonstrated improved biochemical recurrence rates following adjuvant RT in these men. However, the enthusiasm for adjuvant RT remained limited due to the timing of salvage RT in previous trials (which was initiated much later than the contemporary practice), and concerns about the risks of over-treatment and adverse effects of adjuvant RT (such as urinary, bowel, and sexual dysfunction).

The unique design of current meta-analysis can remove some of the biases associated with retrospective meta-analyses and provides an increased number of subjects to allow for sufficiently powered analysis of longer-term survival outcomes. The authors were able to include up-to-date, event-free survival results from 100% of patients randomized in all 3 trials. However, it should be noted that the recurrence events were dominated by biochemical recurrence and not clinical or radiological recurrence, and the overall event (recurrence) rates were low.

The results of additional analyses (which are pre-planned by the authors), of the effects of adjuvant vs early-salvage RT on metastases-free survival, prostate-cancer specific survival, and overall survival are awaited. It’s possible that after longer follow up, the timing of RT and the use of hormone therapy may yield different results in a subgroup with multiple high-risk features. For now, avoiding adjuvant RT (and associated adverse effects) in favor of PSA monitoring and early-salvage therapy appears to be the safest initial option for post-prostatectomy patients with adverse pathology.

1. Vale CL, Fisher D, Kneebone A, et al. Adjuvant or early salvage radiotherapy for the treatment of localised and locally advanced prostate cancer: a prospectively planned systematic review and meta-analysis of aggregate data. 

 Published online September 28, 2020. doi:10.1016/S0140-6736(20)31952-8

PSA monitoring and early salvage therapy appear to be safest in these patients.

Badar M. Mian, MD

Articles

Urologists must remain vigilant regarding post-BCG infection

Extended pelvic lymph node dissection during RP does not appear beneficial

Active surveillance data in prostate cancer show frequent upgrading, rare mortality

Adjuvant RT after prostatectomy does not improve recurrence rate