A Multicenter Retrospective Cohort Series of Muscle-invasive Bladder Cancer Patients Treated with Definitive Concurrent Chemoradiotherapy in Daily Practice

Take Home Message Chemoradiotherapy is a safe and effective treatment option for a wide range of muscle-invasive bladder carcinoma patients with 2-yr locoregional disease-free-survival of 76%. Results in daily practice are comparable with trial outcomes. Lymph node irradiation and carcinoma in situ were not related to outcomes.

CI 42-59%) and 70% (95% CI 62-79%), respectively. On multivariable analysis, cT2 versus cT3-4 tumor stage was significantly associated with better DSS (hazard ratio 1.02, 95% CI 1-1.05, p = 0.024). The 2-yr BI-EFS was 75% (95% CI 69-82%). Forty-three (17%) patients experienced a severe adverse event (grade !3). Limitations include retrospective design and heterogeneous administration of CRT. Conclusions: Concurrent CRT is a safe and effective treatment modality for nonmetastatic MIBC. Patient summary: Chemoradiotherapy for the treatment of muscle-invasive bladder carcinoma is increasingly being applied. In this study, we reviewed the outcomes of this bladder-sparing treatment using a series of patients treated in three hospitals in daily practice. We found that administration of chemoradiotherapy can be safe and effective.

Introduction
Bladder cancer (BC) is among the ten most frequently diagnosed types of cancers for men and women in the Netherlands, with over 6000 new cases in 2019 [1]. A key distinction is made between non-muscle-invasive bladder cancer (NMIBC) versus muscle-invasive bladder cancer (MIBC). While NMIBC can be treated with transurethral resection of a bladder tumor (TURBT) followed by intravesical instillations alone, MIBC requires additional treatment.
Despite treatment, an estimated 50% of patients with nonmetastatic MIBC are alive 5 yr after diagnosis. Treatment guidelines advocate radical cystectomy (RC) plus lymph node dissection ± neoadjuvant chemotherapy (NACT) as a primary treatment option for MIBC, with chemoradiotherapy (CRT) as an alternative for patients who wish to preserve the bladder or for patients not fit for surgery [1]. In 2019, about half of MIBC patients in the Netherlands received RC, with or without NACT and about a quarter received radiotherapy (RT), with or without concurrent chemotherapy (CTx) [2]. Since the BC2001 trial (published in 2012) confirmed the superiority of RT with concurrent CTx for MIBC over RT alone, the focus is slowly shifting to bladder-sparing treatment (BST) through trimodality therapy (TMT) as an alternative for RC [3]. TMT includes maximal TURBT followed by RT and concurrent radiosensitizing CTx. The long-term oncological results in series from large centers and reported in systematic reviews were comparable between RC and TMT [4][5][6][7]. Currently, TMT is offered to well-informed selected patients who opt for BST or for whom RC is not a feasible option.
The aim of this study is to establish oncological and toxicity outcomes in a contemporary series of patients treated with CRT for localized nonmetastatic MIBC in three large centers in the Netherlands.

Patients
This retrospective cohort study was approved by the Institutional

Pretreatment staging and NACT
The three participating hospitals adhered to the European guidelines for MIBC [1]. Patients underwent physical examination, cystoscopy, and CT scan of the chest and abdomen as pretreatment staging. T stage was determined based on CT. The decision to apply TMT was made by a multidisciplinary team.
NACT was generally offered to patients with high-risk features, such as T3-T4a disease, suspicion of PLN metastasis on radiology, or histopathological risk factors on TURBT, such as lymphovascular invasion or variant histology.

Chemoradiotherapy
CRT was administered using volumetric modulated arc therapy or intensity-modulated radiotherapy (IMRT  salvage cystectomy (SC), and (4) death. BI-EFS has been proposed to be a clinically relevant composite outcome measure to assess bladder preservation and oncological safety of BST [9]. Acute toxicity (<90 d of starting CRT) was retrospectively assigned according to the Common Terminology Criteria for Adverse Events (CTCAE) v5.0 [10]. Hematological toxicities were evaluated for the duration of treatment only.

Statistical analysis
Baseline and treatment characteristics, as well as toxicity details, are reported using descriptive statistics.

Overall survival
OS findings are displayed in Figure 2C.  Fig. 2C). No association was found for RT dosage or CTx radiosensitizer (Supplementary Fig. 2D and 2E). NACT was not a predictor of OS (p = 0.55).

Multivariable analyses
Clinical T stage, presence of hydronephrosis at the start of CRT, and radicality of the TURBT were significantly associated with OS and DSS in the univariable analysis (Supplementary Table 1). None of the baseline or treatment parameters, including carcinoma in situ (CIS), hydronephrosis, and elective PLN irradiation, were significantly associated with LDFS. In a multivariable analysis for OS, none of these covariates were significantly associated with OS. In a multivariable analysis for DSS, T3-4 tumors were significantly associated with a higher risk of disease-specific mortality. Furthermore, hydronephrosis lost significance, although a trend for worse OS and DSS was apparent (hazard ratios [HRs] of 1.8 and 1.7, respectively). Results of the multivariable analyses are shown in Table 3.

3.5.
Bladder intact event-free survival BI-EFS findings are displayed in Supplementary Figure 1

Discussion
The aim of the current study was to present an overview of oncological and toxicity outcomes in a contemporary cohort series of patients treated in daily practice with TMT for MIBC in three larger centers in the Netherlands. We report a 2-yr LDFS rate of 76% and 5-yr OS and DSS rates of 50% and 70%, respectively. A higher T stage was negatively correlated with disease-free survival. Of the patients, 7% developed a muscle-invasive recurrence in the bladder. Severe adverse events (grade !3) within 90 d of starting CRT occurred in 17% of patients. BI-EFS at 2-yr follow-up was 75%.    Highest scored  149  61  33  14  8  3  Genitourinary  101  42  23  10  4  2  Frequency  43  18  1  0  0  0  Cystitis, noninfective  46  19  4  2  1  0  Urinary tract infection  7  3  7  3  3  1  Urinary tract obstruction  0  0  3  1  0  0  Urinary retention  5  2  6  3  0  0  Hematuria  3  1  2  1  0  0  Gastrointestinal  44  18  2  1  3  1  Diarrhea  38  16  2  1  0  0  Fistula  0  0  2  1  2  1  Nausea  3  1  0  0  1  0  Obstipation  6  3  1  0  0  0  Hematologi cal  35  14  6  3  1  0  Thrombocytopenia  29  12  3  1  1  0  Thromboembolic event  0  0  3  1  0  0  Miscellaneous  6  3  0  0  0  0  Other  36  15  3  1  0  0  Fatigue  26  The BC2001 trial (2012) was a landmark randomized controlled trial for RT alone versus TMT [3]. In the TMT group, the 2-yr LDFS rate was 67% and the 5-yr OS rate was 48%; these outcomes were superior to those of RT alone. Giacalone et al [4] published the largest singlecenter retrospective series to date, including 475 patients treated from 1986 to 2013 in the Massachusetts General Hospital, over a variety of clinical trials, and reported 5-yr OS and DSS of 57% and 66%, respectively. Mak et al [11] presented a retrospective pooled analysis of six prospective Radiation Therapy Oncology Group studies, and reported 5-yr OS and DSS of 57% and 71%, respectively. The present study shows comparable outcomes. A key distinction between these three studies and our results is performing tumor-site biopsies routinely, which was not a standard procedure in the current study. The results of the current study suggest that this omission does not compromise oncological safety and adds to the feasibility of TMT. In addition, the present study reflects daily practice, including patients who might not have been suitable for the reported clinical studies due to comorbidity and frailty.
Interestingly, only clinical tumor stage proved to be significantly associated with DSS on the multivariable analysis with an HR of 1.02. In previous studies, T3-4 stage and tumor-associated CIS were negatively correlated with survival data [4,12]. Although in the present study, both the presence of hydronephrosis and incomplete TURBT in the univariate analysis were associated with outcome data, this was not the case for tumor-associated CIS-perhaps because patients with predominantly CIS might have been counseled for RC. Nevertheless, our results suggest that patients, including those with limited CIS in combination with MIBC, can effectively be treated with bladder-sparing TMT.
Another debated topic related to CRT treatment is the desirability of elective lymph node irradiation [13]. In a recent consensus meeting of the European Association of Urology and European Society of Medical Oncology, a majority of stakeholders preferred to electively irradiate PLNs in case of CRT [14]. A clinical trial published in 2016, comparing whole-pelvis irradiation with bladder-only irradiation, did not find an improved local control rate with whole pelvis irradiation, although this was associated with higher toxicity than irradiating the bladder only [15]. A confounding factor in whole-pelvis irradiation without elective node target volume is that the generally applied margins of 1-2 cm around the bladder will include the closest lymph nodes. In the present study, the addition of elective PLN irradiation did not impact local control rate or survival over whole bladder only.
Moreover, we found a mild toxicity profile of CRT, favorable to earlier results [3,5]. This could be a result of recent advances in RT techniques, such as IMRT and the use of a simultaneous bladder boost, which reduce radiation on surrounding tissue [16]. Furthermore, the use of capecitabine avoids the need for hospital admission and infusion pumps, adding to the feasibility of CRT [9,17].
Despite the proven safety and feasibility of TMT for MIBC, it is still not widely accepted in clinical practice. Early disappointing results of RT only for BC might have contributed to clinicians' negative opinion of RT-based thera-pies for BC, so that they consequently reserve RT mostly for frail patients unfit for surgery [18,19]. The British SPARE trial, which did not meet recruitment target, has proved that a randomized comparison between BST and RC is not feasible due to patients' and clinicians' preferences [20]. Several systematic reviews have indicated that current literature is biased and provides conflicting results, confirming the lack of an adequate comparison and, therefore, possible slow acceptance [7,21].
Recent advancements with checkpoint inhibitors (CPIs) might accelerate the application of BST. Early results have shown remarkable results of CPIs in combination with RC [22][23][24]. Although no data on the combination of CPIs and TMT are available, several ongoing trials are testing CPIs only and combination of CPIs with CRT, and results are to be expected in the near future [25].
Strengths include the large number of patients, stringent selection criteria, inclusion of cN0 patients and those with pathological tumor-negative lymph nodes, and availability of oncological and toxicity results, thereby providing insight into daily practice of TMT. Limitations are the retrospective nature of the study with its potential underscoring of acute toxicity and the absence of nongenitourinary late toxicity scoring.
To our knowledge, we present the second largest multicenter retrospective cohort of MIBC patients treated with CRT. This study is an extension of the single-center cohort published by Voskuilen et al [9].

Conclusions
Our primary findings reproduce LDFS, DSS, and OS rates in daily practice that are comparable with those of earlier published series, with a low rate of SC performed [4]. Furthermore, this study confirms in the multivariate analysis that patients with T2 versus T3-4 tumors treated with CRT are at a slightly lower risk of disease-specific mortality. In addition, the univariate and multivariate analyses suggest that patients with limited CIS in combination with MIBC can be considered for CRT, and that the addition of elective PLN irradiation does not improve the local control rate or survival outcomes compared with whole bladder only. Moreover, our results provide evidence of safe CRT treatment strategies, with a favorable acute toxicity profile. Other: None.
Financial disclosures: Ben-Max de Ruiter certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.
Funding/Support and role of the sponsor: The work of Ben-Max de Ruiter is funded by the Cure for Cancer foundation and the Dutch Cancer Society.