Introduction
At present, transurethral resection of the bladder tumor (TURBT) combined with postoperative intravesical instillation is the gold standard for the treatment of nonmuscle invasive bladder cancer (NMIBC) (1). However, due to piecemeal resection, traditional TURBT has a high tumor residual rate, making it difficult to provide accurate pathological staging (2, 3). For accurate staging and detection of tumor residue, reresection is recommended for patients with high-risk NMIBC, although it significantly increases the complication risk and financial stress (1, 4).
Different from traditional TURBT, as a new strategy, transurethral en bloc resection of bladder tumor (ERBT) can theoretically wholly remove the bladder tumor and even achieve a 100% detrusor muscle (DM) presence rate. Several recent studies also confirmed that detrusor muscle was present in above 95% of ERBT specimens (5–9). Some previous studies showed that the presence rate of DM was closely related to recurrence and could be a surrogate marker of resection quality (10–12). Although the latest study by Mastroianni et al. showed that the absence of DM has no impact on tumor recurrence, the high DM presence rate and tumor tissue integrity could provide a significant advantage in tumor staging (13, 14). Xu et al. performed reresection on high-risk NMIBC patients who underwent initial ERBT. The results showed that the residual tumor rate and tumor progression rate were only 5.9% and 3.9%, respectively. Moreover, they found that reresection did not seem to improve the prognosis of these patients (5). Given the advantages of ERBT, is it possible to reduce the need for a reresection in high-risk NMIBC patients after initial ERBT?
To answer this question, we conducted a meta-analysis to evaluate the efficacy of ERBT in treating NMIBC by integrating DM presence rate in primary ERBT specimens and tumor residual and upstaging rate in reresection specimens. In addition, we also compared the prognostic indicators of the reresection and non-reresection groups to assess whether patients would benefit from reresection. We believe that if the efficacy of ERBT is satisfactory and the patient cannot derive sufficient benefit from reresection, an attempt can be made to avoid reresection appropriately.
Methods
Search strategy
We conducted research in PubMed, EMBASE, Cochrane Library, and Web of Science up to October 12, 2021, to identify studies on reresection after initial ERBT. The search terms used include: (“bladder neoplasm” OR “bladder cancer” OR “bladder tumor” OR “carcinoma of bladder”) and (“en bloc” OR “en-bloc” OR “en-bloc”) and (“second” OR “repeat” OR “reresection” OR “restaging” OR “reTUR”). We also scanned references of key articles and searched the grey literature to ensure we did not miss any relevant articles. We reported the study according to the preferred reporting items of the systematic review and meta-analysis (PRISMA) (15).
Inclusion and exclusion criteria
Inclusion criteria are as follows: (P) patients diagnosed with primary high-grade Ta (TaHG) or T1 NMIBC who have received initial ERBT; (I) reresection performed within 12 weeks after initial ERBT; (C) no reresection after initial ERBT; (O) outcome indicators should include at least one of the following: detrusor muscle presence rate in primary ERBT specimens, tumor residual rate in reresection specimens, tumor upstaging rate in reresection specimens, comparison of prognostic data between reresection and non-reresection groups; and (S) observational study (prospective or retrospective).
Exclusion criteria are as follows: (a) case reports, comments, conference abstracts, and republished literature; (b) no interest outcome; and (c) data incomplete or invalid.
Selection process and data abstraction
The authors first read the titles and abstracts to conduct a preliminary literature screening. Documents that meet the inclusion and exclusion criteria will be directly included in the full-text evaluation. During the full-text evaluation phase, disputes were settled by two authors through consultation. If no agreement can be reached, a third author was consulted.
Two authors independently extracted data using a predesigned data extraction table. Baseline data included the following: first author and publication year, country, study type, ERBT method, reresection cases, and reresection time. Clinicopathological data included the following: the stage and grade of the primary tumor, primary tumor size, number of primary tumors, location of the residual tumor, follow-up, and prognosis. Data required for meta-analysis included the following: detrusor muscle presence rate in primary ERBT specimens, tumor residual rate in reresection specimens, tumor upstaging rate in reresection specimens, and comparison of prognostic data between reresection and non-reresection groups.
Literature quality and risk of bias assessment
We assessed the quality of literature using a Methodological index for nonrandomized studies (MINORS). The first eight items of MINORS were specially used for quality assessment of noncomparative studies, with 16 points. A score greater than or equal to 12 points was considered moderate to high literature quality (16).
Discussion
To our knowledge, this study is the first meta-analysis to explore whether a reresection can be avoided for high-risk NMIBC patients after initial ERBT. For high-risk NMIBC patients who underwent traditional TURBT, the primary purposes of reresection are to improve the present rate of DM, clarify tumor stage, reduce tumor residue, and improve the prognosis of patients (21, 22). However, our study showed that the present rate of DM in primary ERBT specimens could reach 100%. On this basis, the tumor upstaging rate and tumor residual rate in reresection specimens were extremely low. A recent meta-analysis involving 29 studies also showed that ERBT had a significantly higher DM presence rate in primary ERBT specimens and a significantly lower tumor residual rate in reresection specimens than traditional TURBT. It is consistent with our study (23). In addition, our study also found that reresection did not seem to improve the prognosis of high-risk NMIBC patients with initial ERBT. It can be seen that the advantages of ERBT over traditional TURBT seem to have satisfied the original intention of carrying out reresection. Reresection after initial ERBT in high-risk NMIBC patients does not appear to be critical and essential. Considering the trauma and economic pressure brought by reresection, for patients with poor physical conditions who are difficult to tolerate reresection, it seems that an attempt can be made to avoid reresection appropriately.
When there is no DM in the initial specimen, reresection can provide detrusor muscle of the tumor bed, thus improving the accuracy of tumor staging (24). Gordon's study showed that the present rate of DM in traditional TURBT specimens was 71.2%, which increased to 87.8% after reresection (25). Han et al.'s study showed that the tumor upstaging rate was 16.1% after referring to the reresection specimens (26). A recent systematic review also showed that tumor upstaging occurred in 0%–32% (T1 to ≥T2) of cases (24). In a single-center retrospective study by Zhou et al., DM was present in all 251 ERBT participants’ specimens, and the tumor upstaging rate was only 1.9% (2/108) after reresection of 108 high-risk NMIBC patients (6). Subsequently, Xu et al. also obtained similar results in the study of 115 patients, with the DM presence rate in primary ERBT specimens and the tumor upstaging rate in reresection specimens of 100% and 3.9%, respectively (5). Our study, which integrated all available data, showed that DM was present in 100% of ERBT specimens and the tumor upstaging rate was 0.3% after referring to the reresection specimens. Regarding tumor staging, ERBT has a high presence rate of DM and excellent staging accuracy. Therefore, reresection does not seem to be indispensable in terms of tumor staging.
Cumberbatch et al. conducted a systematic review of studies on reresection after traditional TURT. For Ta tumors, the rate of residual tumors found at reresection ranged from 17% to 67%, and for T1, it ranged from 20% to 71% (24). Subsequently, the study of Akitake et al. also showed that among 143 high-risk NMIBC patients with traditional TURBT, 66 tumor residues (46.2%) were found after reresection (27). Unlike the high tumor residual rate of traditional TURBT, our study showed that patients with initial ERBT found an extremely low tumor residual rate (3.2%) at reresection. In addition, Zhou et al. and Xu et al. performed cystoscopy on patients in the non-reresection group three months after ERBT. They found that the tumor residual rate was similar to that in the reresection group (5, 6). They believe that although the cystoscopy timing differed between groups, the results may have been biased. Nevertheless, in part, it might reflect that reresection after the initial ERBT did not seem to reveal more tumor residuals than non-reresection. In summary, the tumor residual rate of ERBT is low, and reresection may not find more residual tumors. It provides a basis for avoiding reresection.
In a prospective study, patients with T1 NMIBC at initial diagnosis were randomly divided into reresection and non-reresection groups. The first- and third-year recurrence-free survival rates were 82% and 65% in the reresection group and 57% and 37% in the non-reresection group, respectively. It indicates that the reresection can significantly improve the recurrence-free survival rates of patients (28). However, the study of Calo et al. showed that if the initial resection was complete, reresection did not improve RFS and progression-free survival (PFS) in patients with high-grade T1 NMIBC (29). The study of Gontero et al. also pointed out that if the detrusor muscle was not present in the initial TURBT specimen, the RFS and PFS of T1HG patients could be improved by reresection. If the detrusor muscle was present, the patient's prognosis could not be improved by reresection (30). We believe that the mechanism of reresection to improve prognosis lies in removing the DM in the tumor bed and removing the residual tumor as much as possible. In contrast, in ERBT patients who have almost achieved R0 resection, the effect of reresection to improve prognosis will no longer be indispensable. Our results confirm this hypothesis. We found no significant difference in the 1-year RFS rate between the reresection and non-reresection group, nor in the tumor recurrence rate or progression at final follow-up. Due to a lack of data, we included only two studies, which, despite possible bias, have demonstrated to some extent that high-risk NMIBC patients with initial ERBT do not seem to obtain significant improvement in prognosis from reresection.
Limitations
Admittedly, there are still flaws in our research. First, this study is a meta-analysis of the rate and lacks a control group, which cannot directly reflect the difference between ERBT and traditional TURBT. Second, only two studies compared the prognosis of the reresection and non-reresection groups, which is theoretically not suitable for meta-analysis. Third, due to the lack of primary data, we could not detail how many CIS, BCG nonresponse, multifocal, and 3 cm HG bladder tumors were reported in selected studies. Again, because of insufficient data, our study was not limited to T1 cases or included in subgroup analyses. Fourth, we did not consider the possibility of acquiring diabetes in sections far from the deepest part of the tumor, which may have skewed the results. Finally, despite the meta-analysis, the total sample size is still insufficient, and more large-sample randomized controlled studies are needed in the future to verify our results further.