Physician reported treatment patterns and outcomes in metastatic bladder cancer in the USA: the CancerMPact^® Survey 2020

Bladder cancer is the sixth most common cancer in the USA, with an estimated 80,000 new cases and 18,000 deaths in 2020 [1]. Approximately 5% of patients are initially diagnosed with metastatic disease [2]. Additionally, disease recurrence, often resulting in metastatic disease, occurs in half of the patients who undergo cystectomy for early-stage disease [3]. Patients with locally advanced or metastatic bladder cancer have a poor prognosis, with 5-year survival rates of approximately 7.7% [2]. Urothelial carcinoma (UC), which originates in cells that line the urinary tract, accounts for >90% of bladder cancer cases [3].

Advanced bladder cancer is considered a chemotherapy-sensitive tumor, and standard treatment has been based on platinum-containing chemotherapy since the publication of pivotal phase II/III trial data more than 20 years ago [4,5]. According to the US National Comprehensive Cancer Network (NCCN) Guidelines, cisplatin- or carboplatin-containing treatment is recommended for patients who are eligible or ineligible for cisplatin, respectively [3]. Recommended first-line (1L) regimens for cisplatin-eligible patients (category 1 recommendation) are ≤6 cycles of cisplatin plus gemcitabine or a dose-dense combination of methotrexate, vinblastine, doxorubicin and cisplatin (MVAC) [6,7]. For patients who are cisplatin ineligible, perhaps due to renal impairment, poor performance status, hearing loss, peripheral neuropathy, symptomatic heart failure or other comorbidities, the recommended 1L chemotherapy is carboplatin plus gemcitabine [3,8,9].

UC is an immunogenic tumor, which provided the rationale for assessing immune checkpoint inhibitors (ICI) in this disease [3,10–13]. Based on single-arm, phase II trials, anti–PD-L1 (atezolizumab) and anti–PD-1 (pembrolizumab) antibodies were granted accelerated approval in 2017 for 1L treatment of cisplatin-ineligible patients with advanced UC [3,14,15]. However, in 2018, 1L approvals for both agents were restricted to cisplatin-ineligible patients with PD-L1–positive tumors or platinum-ineligible patients [16–18]. Additionally, in 2021, the approval of pembrolizumab was converted from accelerated to a full approval, but its 1L label was further restricted to platinum-ineligible patients only [19]; this restriction was based on results of a randomized phase III trial that did not demonstrate any significant efficacy benefit for 1L pembrolizumab monotherapy over chemotherapy in cisplatin-eligible patients [16,20]. Previously, in 2016 and 2017, atezolizumab, avelumab, durvalumab, nivolumab and pembrolizumab as monotherapies were approved in the US as second-line (2L) treatments for patients with advanced UC whose disease progressed following platinum-containing chemotherapy (or within 12 months of neoadjuvant or adjuvant treatment with platinum-containing chemotherapy) and became NCCN preferred treatments in this setting. However, accelerated US approvals of atezolizumab and durvalumab were withdrawn in 2021 following the failure of confirmatory trials [3,18].

In June 2020, the landscape for 1L treatment of advanced UC in the USA changed drastically following the US FDA approval of avelumab, an anti–PD-L1 monoclonal antibody, as 1L maintenance treatment for patients with advanced UC without disease progression following 1L platinum chemotherapy. This approval was based on results from a randomized phase III trial (JAVELIN Bladder 100; NCT02603432) that showed significantly prolonged overall survival in the avelumab plus best supportive care arm compared with best supportive care alone [21]. Subsequently, avelumab 1L maintenance was added to NCCN guidelines as a preferred treatment for cisplatin-eligible or -ineligible patients with UC without disease progression following 1L platinum chemotherapy, irrespective of PD-L1 status (category 1 recommendation) [3]. In addition to ICIs, two antibody–drug conjugates, enfortumab vedotin and sacituzumab govitecan, have been FDA approved for patients who have disease progression after treatment with both platinum-containing chemotherapy and ICI treatment (both agents), as well as for those cisplatin-ineligible patients who have received ≥1 prior line of therapy (enfortumab vedotin only). Also, an FGFR inhibitor, erdafitinib, has been FDA approved for patients with susceptible FGFR alterations who have disease progression after platinum-containing chemotherapy [22–24]. An overview of current treatment guidelines for advanced UC is provided in Supplementary Figure 1.

Because of these major changes in the treatment landscape for advanced bladder cancer in recent years, it is important to understand real-world treatment patterns and outcomes before and after the integration of new treatment options so that their impact can be assessed. This paper presents the results of a quantitative analysis assessing the clinical management of metastatic bladder cancer as reported by US physicians using the CancerMPact^® (CMP) survey. The survey was developed after the approval of 1L treatment with ICI monotherapy in selected subgroups but before the approval of avelumab in 1L maintenance (on 30 June 2020) and other agents approved more recently; thus, questions related to avelumab 1L maintenance treatment were not included.

Materials & methods

Procedures & assessments

Physicians were recruited to participate in the online CMP survey fielded in July 2020. Eligible physicians were board-certified oncologists or urologists who had ≥3 years of clinical experience and were treating ≥10 patients with any stage of bladder cancer per month at the time of the survey. Although the CMP covered questions regarding treatment for all disease stages, this manuscript focuses on physician-reported outcomes in patients with metastatic stage IV (M1) bladder cancer (Supplementary Figure 2).

Physicians were asked about their experience and medical practice (e.g., years in practice, type of practice, practice specialty, patient volume, etc), and about treatments administered to their patients within the last 6 months, including lines of therapy administered, therapy modality (e.g., radiation, surgery, antibody, targeted therapy, chemotherapy, etc) and systemic therapy regimens. Additionally, for patients with metastatic disease, physicians were asked about 1L therapy utilization by cisplatin eligibility. For most questions, physicians were asked to consider patients they had treated in the last 6 months and to report, based on recall, the proportion of patients in their practice with respect to a specific treatment, testing or outcome. However, for questions about the proportion of patients moving to subsequent lines of systemic therapy, physicians were asked to consider all of their patients with metastatic bladder cancer regardless of time period. Definitions for line of therapy, recurrence, cisplatin eligibility, response to therapy and long-term response were not provided in the survey, and physicians responded to all survey questions based on their own clinical practice and experience.

The online survey is part of the CMP proprietary data source from Cerner Enviza comprising data on cancer epidemiology and clinical management strategies. The survey is updated annually to capture changes in treatment practice. To ensure the annual survey reflects current practices and potential changes in treatment landscape, the Cerner Enviza internal research project team reviews treatment guidelines and regulatory approvals; they also conduct a targeted literature review of pivotal trial data published in peer-reviewed medical journals and at major oncology congresses prior to each annual survey collection [3].

Data collection & analysis

The physician survey was programmed, fielded and hosted by Kantar Profiles; this is a large, verified online panel where healthcare professionals that are recruited via various channels to participate in online surveys. Anonymized raw data were transferred securely to Cerner Enviza, where personnel performed the tabulation and data analysis. For each survey, two analysts independently reviewed the data file. Where data were outlying or incomplete, data were excluded as needed. All results from the survey relating to patients and treatment were reported as an unweighted mean of all responses from physicians. Subgroups defined by physician specialty (urology vs oncology) and by practice setting (academic vs nonacademic) were also analyzed. Data were analyzed descriptively, and no formal statistical analysis was performed.

This CMP survey did not collect or use patient-level data or any other data involving people, medical records or human samples. All information was retrieved from online physician surveys that obtained information around overall treatment patterns only. Therefore, no institutional review board approval was necessary.

Results

Survey population

In total, 106 USA-based physicians participated in the CMP survey (Table 1). At the time of the survey, the responding physicians reported having on average 16.7 years (range: 3–33 years) of practice after residency and treating a mean of 35 patients (range: 10–160 patients) per month with any stage of bladder cancer. Oncologists accounted for 57.6% of physicians, and urologists accounted for 42.5%. Most physicians practiced within a community setting (75.4%), and a smaller proportion practiced in an academic setting (25.0%).

Responding physicians reported that, on average, 26.8% of their patients had an initial diagnosis of stage IV metastatic bladder cancer; 30.9% of the patients they were treating at the time of the survey had metastatic bladder cancer (Supplementary Table 1). Oncologists reported treating substantially more patients with stage IV disease than urologists (62.9 vs 12.5%; Supplementary Table 2). Physicians in an academic and nonacademic setting treated similar proportions of patients with each stage of disease stage (Supplementary Table 3). Furthermore, 21–32% of patients with nonmetastatic bladder cancer (stage I–IV M0) who responded to initial therapy had disease recurrence within 1 year. Metastatic recurrence was rare for patients with stage I disease (7.4%) but was common for more advanced stages (25.5% for stage II, 35.9% for stage III and 40.5% for stage IV M0; Figure 1).

Treatment modalities

Of patients receiving initial treatment for metastatic bladder cancer, the most commonly reported treatment modality was systemic therapy (86.3%), which was administered alone (62.0%) or in combination with cystectomy (7.1%), radiotherapy (8.7%) or both cystectomy and radiotherapy (8.5%; Supplementary Table 4). A small proportion of patients were reported to receive radiation and/or cystectomy only (9.6%) or received no therapy (4.1%). Oncologists reported prescribing systemic therapy to more patients than urologists (89.6 vs 61.4%; Supplementary Table 5).

Patients receiving 1L systemic therapy

Physicians reported that the mean proportions of patients with metastatic bladder cancer who were cisplatin eligible or ineligible were 59.9 and 40.1%, respectively (Supplementary Figure 3).

In cisplatin-eligible patients, physicians reported that a majority received platinum-containing chemotherapy (86.1%). The most frequently used systemic therapy regimen in cisplatin-eligible patients was cisplatin plus gemcitabine in 47.3% (Table 2). The second most frequently reported systemic therapy was MVAC, received by a mean of 10.3% of patients. First-line ICI treatment (including atezolizumab, avelumab, durvalumab, nivolumab and pembrolizumab) was reported for 9.1% of cisplatin-eligible patients. Urologists reported prescribing ICI treatment to more cisplatin-eligible patients than oncologists (16.0 vs 9.1%; Supplementary Table 6).

For cisplatin-ineligible patients, the most frequently used regimen was carboplatin plus gemcitabine in 40.3% of patients, followed by pembrolizumab in 22.6% and atezolizumab in 14.2%. Physicians practicing in an academic setting reported prescribing ICI treatments to 56.3% of cisplatin-ineligible patients, whereas physicians practicing in a nonacademic setting prescribed ICI treatments to 43.2% of cisplatin-ineligible patients (Supplementary Table 7).

Subsequent lines of systemic therapy

Of the 106 physicians surveyed, 67 provided responses regarding subsequent therapy. Of cisplatin-eligible and -ineligible patients, 43.3 and 44.1%, respectively, were estimated to have disease progression following 1L systemic therapy and received 2L therapy (Table 3). Proportions estimated to receive 2L therapy after disease progression reported by oncologists were higher than proportions reported by urologists for both cisplatin-eligible patients (47.6 vs 11.9%) and cisplatin-ineligible patients (47.9 vs 17.5%; Supplementary Table 8). The distribution of patients who received 2L and subsequent lines of systemic treatment by physician setting are also presented in Supplementary Table 9. Overall, ICIs accounted for 62.2% of 2L therapies reported, with pembrolizumab (25.3%) and nivolumab (14.9%) most commonly used (Supplementary Table 10). Oncologists reported prescribing 2L ICI treatment to a higher proportion of patients than urologists (64.4 vs 40.8%; Supplementary Table 11). In response to questions about timing of 2L treatment initiation from the start of 1L treatment, physicians reported that approximately half of patients initiated 2L treatment within 12 months, and approximately one third of patients initiated 2L treatment after 13–24 months (Supplementary Table 12). Oncologists reported that a higher proportion of patients initiated 2L treatment within 12 months than urologists (45.5 vs 26.0% of cisplatin-eligible patients; 48.9 vs 33.3% of cisplatin-ineligible patients; Supplementary Table 13). The timing between 1L and 2L treatment did not seem to differ between physicians practicing in an academic or nonacademic setting (Supplementary Table 14). In addition, 16.1% of cisplatin-eligible and 18.0% of cisplatin-ineligible patients had disease progression and were alive at the time of the survey but were not receiving 2L therapy due to patient choice, comorbidities, age or cost (Table 3). Furthermore, physicians reported that 18.0 and 20.8% of cisplatin-eligible and -ineligible patients, respectively, did not achieve a long-term response to 1L therapy and died prior to receiving 2L therapy. However, 22.6% of cisplatin-eligible patients and 17.1% of cisplatin-ineligible patients achieved a long-term response to 1L therapy and did not receive 2L therapy; reasons for not receiving 2L were not specified.

In response to questions about third-line (3L) or fourth-line (4L) therapy, 63 and 57 physicians provided responses, respectively, and data were combined for cisplatin-eligible and -ineligible patients. Of patients who received 2L therapy, an estimated 34.6% had disease progression and received 3L therapy, 21.5% had disease progression and were alive but did not receive 3L therapy, 26.0% did not have a long-term response to 2L therapy and died, and 17.9% had a long-term response to 2L therapy and did not receive 3L therapy (Table 3). Among patients who received 3L therapy, 37.4% of patients received an ICI, most commonly pembrolizumab.

Outcomes by line of therapy

Among cisplatin-eligible and -ineligible patients receiving 1L therapy, respectively, a complete response (CR) was estimated to occur in 22.9 and 16.7% of patients, and partial response (PR) was reported in 34.7 and 32.1% (Table 4). Among patients receiving 2L or 3L therapy, respectively, CR was reported in 11.5 and 8.8%, and PR was reported in 25.1 and 18.9%. The overall treatment pathway across all lines of therapy is summarized in Supplementary Figure 3. Oncologists and urologists reported similar proportions of patients with a CR, a PR, stable disease (SD) and progressive disease by lines of therapy received (Supplementary Table 15). Physicians practicing in an academic versus a nonacademic setting reported higher rates of CR following 2L and 3L therapy (2L: 19.4 vs 9.5%; 3L: 15.4 vs 7.0%; Supplementary Table 16).

Discussion

We report the results of a nationwide survey of USA-based physicians that obtained detailed information about treatment patterns in patients with metastatic bladder cancer. These data regarding the treatment landscape for metastatic bladder cancer in the USA, generated in July 2020, reflect the initial introduction of ICI monotherapy into clinical practice as 1L therapy for cisplatin-ineligible patients and as 2L therapy following disease progression with platinum-containing chemotherapy. Physicians reported that 26.8% of their patients had metastatic stage IV disease at initial diagnosis, which is much higher than the proportion reported by the Surveillance, Epidemiology, and End Results registry (5%) [2]; however, this discrepancy may be due to the majority of physicians surveyed being specialized oncologists, who are most likely to treat patients with later-stage disease.

Physicians reported that the most commonly administered type of 1L therapy for both cisplatin-eligible and -ineligible patients was platinum-containing chemotherapy. Approximately two-thirds of cisplatin-eligible patients received either cisplatin plus gemcitabine or MVAC, and approximately 40% of cisplatin-ineligible patients received carboplatin plus gemcitabine. These findings demonstrate that physicians are largely adherent to NCCN guidelines, which recommend the use of 1L gemcitabine with either cisplatin or carboplatin, dependent on cisplatin eligibility [3]. Approximately half of cisplatin-ineligible patients received 1L ICIs, which are NCCN-recommended treatment options for a subset of these patients [3].

In previous clinical studies in advanced UC, 1L platinum-containing chemotherapy resulted in objective responses in approximately 40–49% of patients and SD in approximately 30% [4,25]; these rates are similar to those of real-world, retrospective studies [26–28]. In our survey, outcomes were not reported by treatment; however, response rates across all 1L treatments were generally consistent with previous reports, including CR or PR in 49–58% of patients and SD in 22–25% of patients [4,5,25]. Physicians reported that only 17.1–22.6% obtained a long-term response to 1L therapy and that 18.0–20.8% died before receiving 2L therapy. These data highlight the significant need to improve long-term outcomes with 1L therapy in patients with metastatic bladder cancer and prolong the benefits of response or SD achieved with 1L treatment modalities. Physicians also reported that approximately 44% of patients received 2L therapy after 1L treatment, which is slightly higher than rates reported in previous real-world studies [27–34]. Approximately half of patients receiving 2L systemic therapy received ICIs, with pembrolizumab and atezolizumab being the most frequently administered agents, consistent with NCCN guidelines at the time of the study.

The results of this study provide an important benchmark for understanding the treatment landscape of metastatic bladder cancer, and specifically real-world treatment patterns and outcomes as reported by treating physicians, following the approval of 1L ICI therapy and before the approval of avelumab 1L maintenance treatment. Few published studies have reported on contemporary treatment patterns and outcomes in patients with metastatic bladder cancer by line of therapy and eligibility for cisplatin. A systematic review by Hepp et al. highlighted the lack of up-to-date country specific information on treatment patterns for patients with advanced or metastatic bladder cancer [35]. Thus, this study supports assessment of the evolving treatment landscape of metastatic bladder cancer.

Results from this study should be also interpreted in the context of its limitations. First, the study is reliant on responses to a survey, physicians’ interpretations of clinical characteristics and estimated outcomes (e.g., cisplatin eligibility, lines of therapy, assessments of CR, PR and SD, long term response, etc) and has a cross-sectional design. Additionally, physicians provided answers based upon the patient pool they were treating at the time of the study, which may or may not be representative of the larger population of patients with metastatic bladder cancer. As noted earlier, definitions of line of therapy, recurrence, cisplatin eligibility and response to therapy were based on physicians’ own experience. The CMP survey did not include questions regarding platinum ineligibility, so it was unclear whether patients who were considered cisplatin ineligible were also ineligible for any platinum-containing chemotherapies (i.e., carboplatin). Patient characteristics and treatment outcomes were provided by physicians based on recall, without reviewing individual patient charts; thus, responses were prone to recall bias. However, the survey limited the recall to the last 6 months for most questions to reduce recall bias. The study did not include patient-level data to validate or determine the reasons behind physicians’ decisions. Physicians were asked about lines of therapy but not about sequence of treatment, and differentiation of outcomes by treatment type was not possible. In addition, percentage values were reported as unweighted means of physicians’ responses, which provide an overall picture but cannot be evaluated via formal statistical analysis. Lastly, the sample size for urologists was small. Despite these limitations, screening criteria for the survey were designed to ensure that participating physicians had substantial experience in the treatment of metastatic bladder cancer and that physicians who provided responses are representative of UC care in the USA, with the inclusion of physicians working in community practice and academic settings, as well as both oncologists and urologists. Thus, these responses provide valuable insights into physicians’ observations and decisions in a US clinical practice setting.

Shortly before the survey was conducted, avelumab received FDA approval as 1L maintenance therapy for patients with advanced UC that has not progressed with 1L platinum-containing chemotherapy. Consequently, survey questions did not ask questions regarding the uptake of avelumab 1L maintenance. The approval of avelumab 1L maintenance was based on results from the pivotal phase III JAVELIN Bladder 100 trial, which also supported the inclusion of the JAVELIN Bladder regimen in international guidelines as a preferred approach (category 1 recommendation) [3,21,36,37]. In contrast, phase III trials of atezolizumab (IMvigor130), pembrolizumab (KEYNOTE-361) and durvalumab (DANUBE) in the 1L setting, which included treatment arms receiving ICI monotherapy or ICI plus chemotherapy (IMvigor130 and KEYNOTE-361) or anti–CTLA-4 combinations (DANUBE), did not show improved overall survival versus chemotherapy alone [20,38,39]. As previously mentioned, this survey was conducted prior to the FDA withdrawals of the approvals for durvalumab and atezolizumab for 2L treatment of metastatic UC and the amended FDA approval of pembrolizumab for 1L treatment of metastatic UC. Furthermore, later-line treatment options have since expanded following the approvals of enfortumab vedotin, sacituzumab govitecan and erdafitinib [22–24]. Therefore, future real-world studies are needed to evaluate the changing treatment landscape and the impact of avelumab 1L maintenance on the treatment paradigm.

Conclusion

This survey, completed by US-based physicians, provided valuable insights about treatment patterns and outcomes in patients with metastatic bladder cancer in 2020, which followed the approvals of ICI monotherapies, but was before the approval of avelumab 1L maintenance and readouts from phase III trials of other ICI monotherapies in the 1L UC setting. Platinum-containing chemotherapy was the most widely reported 1L treatment approach for both cisplatin-eligible and -ineligible patients, with cisplatin- and carboplatin-containing regimens being most commonly prescribed, consistent with international guidelines. Physician-reported treatment outcomes appeared to be consistent with previous clinical trials and real-world studies, including the limited long-term benefits with 1L therapy and the high proportion of patients who did not receive 2L therapy or later lines, which emphasizes the need for improved treatment options. Future research is warranted to investigate the impact of emerging trial findings and newer treatment options, including avelumab 1L maintenance, on clinical practice and patient outcomes.