Bladder cancer is one of the most common urinary tract tumors in the US, with an estimated 80,500 new cases and 17,600 deaths in 2019 (1). Patients with advanced and metastatic bladder cancer had lower survival rates, with 5-year survival rates of 34% in localized disease, 7% in regional disease, and 5% in metastatic disease.

Radical cystectomy (RC) plus pelvic lymph node dissection is commonly regarded as the gold standard therapy for muscle-invasive bladder cancer (MIBC) (2–4). But some patients have a strong willingness to preserve their own bladders, and bladder-sparing becomes their preferred option. The existing bladder preservation approaches are partial cystectomy, transurethral resection (TUR) alone, single chemotherapy, or radiation therapy (RT). It is generally considered that monotherapy is inferior to RC for MIBC.

Recently, a trimodal therapy (TMT), which includes utmost TUR followed by simultaneous chemotherapy and RT, is the most effective strategy for preserving the bladder (5, 6). Several studies reported that for overall survival (OS), the effects of TMT on MIBC were parallel to RC (7, 8). A published series systematic review indicated that TMT results in satisfactory outcomes and may be a rational therapy option in well-selected patients (9). Furthermore, several clinical trials compared the outcomes between RC and TMT (10–12). Most of the literature included in the previous meta-analysis were case series (9, 13, 14) and did not directly compare the efficacy of RC and TMT; and not all included patients were performed standard TMT in the other meta-analysis (15). So, it is essential to perform a meta-analysis directly comparing RC and standard TMT. The aim of this meta-analysis was to evaluate the difference in OS and other outcome indicators after using either of the two treatment modalities in MIBC because we believe that accumulating evidence from studies should be more reliable.

Since there is a lack of RCTs directly comparing RC and bladder preservation therapy with TMT, and the previous systematic review mainly focused on case series or incomplete TMT, this study uses the meta-analysis method to analyze the effects of both complete TMT and RC on MIBC from available clinical controlled trials. The data suggest that compared to complete TMT patients, RC patients have a higher overall OS, CSS, and less mortality for MIBC patients. According to the stratified analysis, it was found that there was a statistically significant difference at more than 10-year OS between the two groups. Moreover, the results demonstrated that the bladder cancer patients belonging to CCS “0” score preferred RC, while CCS “2” score's patients were prone to TMT.

Previous meta-analysis studies have shown inconsistencies for TMT and RC. Arcangeli et al. (14) reported that TMT can generate outstanding 5-year OS rates between these two interventions. Fahmy et al. (13) indicated that the OS and disease-specific survival (DSS) were comparable between TMT and RC: the average 10-year OS rate was 30.9% for TMT and 35.1% for RC, respectively (P = 0.32), and the average 10-year DSS rate was 50.9% for TMT and 57.8% for RC, respectively (P = 0.26). García-Perdomo et al. (15) reported that CSS rate favored patients who underwent RC. However, most included studies were case series in the systematic reviews by Arcangeli et al. (14) and Fahmy et al. (13). In the study by García-Perdomo et al. (15), some studies included only chemoradiotherapy that was not the standard TMT. Ploussard et al. (9) only included patients receiving TMTs.

It is reported that the 5-year OS was from 50 to 60% in the literature. The Radiotherapy Oncology Group (RTOG) has completed six prospective TMT regimens for MIBC patients undergoing cystectomy. Five of the RTOG regimens are phase I–II trials for simultaneous chemotherapy and radiotherapy, and one is a phase III trial to test the efficacy of adjuvant chemotherapy with methotrexate, cisplatin, and vinblastine. A total of 415 patients were entered on these trials. The 5-year OS rate was near to 50%, with 75% of those patients accomplishing a cure for their bladder cancer while maintaining bladder function (23). In the most recently published long-term follow-up of 348 patients from the Massachusetts General Hospital (MGH) in the USA, they show that 5-year OS rates and CSS rates were 52 and 64% for split-course TMT, respectively (24). Another continuous-course TMT series comes from Erlangen, Germany. A total of 331 patients were permanently followed up, and the overall 5-year OS rate was 54% (25). Compared to TMT, RC has been reported in previous studies to show comparable outcomes. These studies have reported that 5-year OS rates were 58–68% in patients receiving primary RC (26–28). Our meta-analysis showed that the longest follow-up time was <10 years in three studies, and no difference was found between them at <10 years' OS and CSS.

However, the final follow-up time was more than 10 years in the five studies; RC was superior to TMT at more than 10 years' OS and CSS. Faraj et al. (29) reported that the 10-year OS was 45.6% for RC patients. For bladder-preserving combined-modality therapy, the 10-year OS rates were 36–39% (7, 30).

The adverse effects of RC were associated with sexual dysfunction and external urinary drainage devices, which cause emotional and psychological stress. However, two prospective trials have confirmed that bladder-sparing must be beneficial to improve quality of life (QOL) after TMT compared to RC (31, 32). TMT was reported to be associated with better sexual function (P < 0.02) and better body image perception (P < 0.001) compared to RC; TMT had better general QOL compared with those who had received RC (P = 0.001) and higher physical, role, social, emotional, and cognitive functioning (P ≤ 0.04); TMT was associated with better bowel function (P = 0.02) and fewer bowel symptoms (P ≤ 0.05) (33).

Although the patients are afraid of TMT's toxicity, studies show that it is acceptable for patients with bladder preservation treatment. The main acute toxicities include hematologic, gastrointestinal (GI), and genitourinary (GU). In the Bladder Cancer 2001 trial (6), 182 patients underwent chemoradiotherapy and median follow-up was 69.9 months; the results indicated that there was slightly increased acute grade 3 or 4 adverse events (AEs) in the chemoradiotherapy group, and these events were mostly GI toxic effects.

The proportion of late grade 1–2 toxicity was from 6 to 25% for GU and 5 to 6% for GI toxicities, respectively (34–37). Among which, the primary low-grade toxicities included urgency, nocturia, dysuresia, incontinence, hemorrhagic cystitis, diarrhea, and rectitis. The percentage of late grade 3 urinary tract toxicity was from 3 to 8% of the series (6, 38, 39). In the BC2001 trial, there were no differences in late toxicity after adding chemotherapy to RT; of these, the grade 3–4 toxicity rate was 0.8% for GI symptoms and 7.4% for GU symptoms, respectively (6).

Based on these studies, it is clear that TMT is safe and effective for bladder cancer patients without serious side effects. However, we should note that TMT generates higher treatment costs, which is estimated at an excess spending of $468 million within 1 year of diagnosis in the US (40). Therefore, we should adopt multidisciplinary consultation with experts and also consider the patient's treatment expectations and financial circumstances, which may make bladder cancer patients access optimal treatment.

This study has limitations due to the retrospective design of the included cases, which is easy to be affected by selection bias. Firstly, the lack of studies assessing and reporting the stage-based outcomes made it difficult to evaluate any obvious difference in survival between TMT and RC in the subgroups of patients with different stages such as T2 or >T2 tumor stage. Secondly, neoadjuvant chemotherapy may influence overall survival regardless of whether patients undergo RC or TMT. Thirdly, the papers abstracted from the database have obvious heterogeneity in their treatment methods, such as open or minimum invasive surgery for RC, and greatly differ from treatment regimen, such as radiation dose and fraction, field of irradiation, chemo agent (cisplatin use or not, combination of agent), etc. Especially, the effect of chemotherapy maybe quite different between intra-arterial infusion and systemic administration of chemo agent. So, the optimum radiation exposure and techniques in TMT need to be further investigated. Fourthly, there may be a language bias, as all included articles were published in English. Fifthly, some outcomes (any grade AEs, dose reductions) had significant heterogeneity, but they may have influenced the results. Moreover, data from the included studies were analyzed using propensity score matching to weaken the impact of treatment selection bias and potential confounding factors that are often faced in observational studies. We should note that compared to the RC group, the patients' age receiving TMT was older; therefore, the results should be interpreted carefully. Although lacking of RCTs, we believe that this meta-analysis provides valuable information for patients and clinicians.

Overall, this meta-analysis demonstrates that the efficacy of TMT is non-inferior to that of RC at <10-year OS, and RC is superior to TMT at more than 10-year OS. Therefore, TMT may be a reasonable treatment option in well-selected patients who are unsuitable for surgery or are not willing to experience surgery. In the future, more high-quality, large-sample RCTs are needed to verify the results.

All datasets presented in this study are included in the article/ supplementary material.

HD: conceived the study, participated in its design, and coordinated and drafted the manuscript. NF and DM collected the data. HD, NF, ZN, and DM performed the statistical analysis. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.