Abstract
Background
Patients with bladder cancer are apt to develop multiple recurrences that require intervention. We examined the recurrence, progression and bladder cancer-related mortality rates in a cohort of individuals with high-grade non-muscle-invasive bladder cancer.
Methods
Using linked SEER-Medicare data, we identified subjects with a diagnosis of high-grade, non-muscle-invasive disease in 1992–2002 and were followed until 2007. We then used multivariate competing-risks regression analyses to examine recurrence, progression, and bladder cancer-related mortality rates.
Results
Of 7,410 subjects, 2,897 (39.1%) experienced a recurrence without progression, 2,449 (33.0%) experienced disease progression, of whom 981 succumbed to bladder cancer. Using competing-risks regression analysis, we found the 10-year recurrence, progression, and bladder cancer-related mortality rates to be 74.3%, 33.3%, and 12.3%, respectively. Stage T1 was the only variable associated with a higher rate of recurrence. Women, black race, undifferentiated grade, stage Tis and T1 were associated with a higher risk of progression and mortality. Advanced age (≥70) was associated with a higher risk of bladder cancer-related mortality.
Conclusions
Nearly three-fourths of patients diagnosed with high-risk bladder cancer will recur, progress, or die within ten years of their diagnosis. Even though most patients do not die of bladder cancer, the vast majority endures the morbidity of recurrence and progression of their cancer. Increasing efforts should be made to offer patients intravesical therapy with the goal of minimizing the incidence of recurrences. Furthermore, the high recurrence rate seen during the first two years of diagnosis warrants an intense surveillance schedule.
Keywords: Urinary Bladder Neoplasms, Recurrence, Progression, Bladder Cancer Mortality, Quality of Healthcare
INTRODUCTION
Over the last 15 years, bladder cancer-related mortality has fallen by only 5%, while cancer-specific mortality rates have fallen more precipitously for prostate, breast, lung and colon cancer.1 This may in part be attributed to quality of care. While overutilization of health care services and the attendant waste of scarce health care resources have generated much attention, underutilization of effective (and reimbursed) care have not. For instance, in patients with muscle-invasive disease who underwent radical cystectomy, 9% received neoadjuvant chemotherapy,2 35% were not treated within three months of diagnosis,3 and among subjects enrolled in a randomized control trial involving neoadjuvant chemotherapy and radical cystectomy, 45% did not receive a standard lymph node dissection.4 Suboptimal bladder cancer care isn't just limited to those with more advanced disease; it transcends stage and is just as prevalent in non-muscle-invasive disease. Madeb et al. found that only 0.3% of subjects undergoing endoscopic resection received perioperative intravesical chemotherapy,5 and Huang et al. found that 58% of patients with high-grade, non-muscle-invasive bladder cancer did not receive even a single instillation of immunotherapy or chemotherapy within one year of diagnosis.6 In fact, in a national, claims-based sample of 4,545 cases with high-grade, non-muscle-invasive bladder cancer, we discovered that only one individual received all the follow-up surveillance and treatment measures as recommended by the American Urological Association (AUA) and National Comprehensive Cancer Network (NCCN).7 Moreover, patients who received at least half of the recommended surveillance and treatment measures enjoyed a statistically significant survival advantage—a process-outcomes link.8
In the face of our earlier findings about underutilization of care for patients with bladder cancer, some have posited that the surveillance and treatment strategies for patients with high-grade non-muscle-invasive bladder cancer, as prescribed by the national guidelines, are unnecessarily intensive, thus resulting in apparent non-compliance. Until now, the natural history of this disease in the US has not been described in population-based data. In this context, we sought to more fully characterize the natural history of the disease—recurrence, progression, and bladder cancer-related mortality rates—using linked Surveillance, Epidemiology and End Results (SEER)-Medicare database. We hypothesized that the vast majority would have recurrences and require interventions.
METHODS
Data Source
We used the SEER-Medicare-linked database of the National Cancer Institute, which contains data on individuals aged 65 years and older, to identify bladder cancer patients diagnosed in 1992–2002. SEER data are summarized in the Patient Entitlement and Diagnosis Summary File (PEDSF) and contains data on patient demographics, tumor characteristics and follow-up information. The PEDSF was linked with 100% of Medicare claims from inpatient, outpatient, and national claims history files and was restricted to subjects who had Medicare Fee-for-Service coverage and for whom Medicare Parts A and B claims data were available for 12 months prior to diagnosis of bladder cancer.9
Study Population
The cohort consisted of patients at least 66 years of age with an incident diagnosis of non-metastatic (N0M0), high-grade (poorly or undifferentiated) urothelial (histology codes 8120, 8130), non-muscle-invasive (Ta, Tis, T1) bladder cancer (International Classification of Diseases, Ninth Revision (ICD-9) 188.0–188.9, 233.7) diagnosed between January 1, 1992 and December 31, 2002, for whom claims data were available through 2007. While beneficiaries are eligible for Medicare coverage at age 65, we limited our cohort to those 66 or older to allow at least one year of eligibility in Medicare before bladder cancer diagnosis to ascertain comorbidity data. We restricted our analysis to those with known grade, stage and histology.
Study Variables
From the PEDSF, we determined patient age (66–69, 70–74, 75–79, ≥80), gender, race/ethnicity (White, Black, Hispanic, Other), marital status (married, other), tumor grade (poorly differentiated, undifferentiated), T-stage (Ta, Tis, T1), and year of diagnosis. We imputed subject socioeconomic status by utilizing 2000 US Census data to derive quartiles of ZIP code-level median household income (<$35,000, $35,000–$45,000, $45,001–$55,000, >$55,000) and percent of residents ≥25 years of age with at least four years of college education (categorical: <15.0%, 15.0%–25.0%, 25.1%–35.0%, >35.0%).10 We used the Klabunde et al modification of the Charlson Comorbidity index to quantify severity of preexisting comorbidities (0, 1, 2, ≥3).11, 12 For each patient, we noted the provider and institution where the initial bladder cancer was diagnosed utilizing the Unique Physician Identifier Number (UPIN) and the corresponding institution (Medicare provider number). We defined recurrence as an inclusive state that includes 1) recurrence without progression (receipt of a transurethral resection of a bladder tumor); 2) progression (receipt of radical cystectomy, radiotherapy, or systemic chemotherapy); or 3) bladder cancer-related death beyond 90 days of diagnosis. Progression was defined as receipt of radical cystectomy, radiotherapy, systemic chemotherapy, or bladder cancer-related death beyond 90 days of diagnosis.
Statistical Analysis
Since patients may die from non-cancer causes, we utilized a maximum likelihood, competing-risks regression model as described by Fine and Gray to determine recurrence, progression, and bladder cancer-related mortality rates.13 We performed competing-risks regression analyses to characterize the incidence of recurrence, progression, and cancer-specific mortality. For recurrence, we defined the event of interest as recurrence, progression, or bladder cancer-related death. For progression, we defined the event of interest as progression or bladder cancer-related death. For mortality, we defined the event of interest as bladder cancer-related death. The competing event was defined as non-cancer-related mortality for all scenarios—recurrence, progression, and cancer-specific mortality. This model adjusted for patient age, gender, race/ethnicity, marital status, ZIP code-level income and education, comorbidity, and tumor grade and stage. Estimates are reported as sub-hazard ratios (HR) with corresponding 95% confidence intervals. Additionally, since patients treated by the same provider may have similar outcomes, we accounted for potential clustering by utilizing the Huber-White sandwich variance estimator to the competing-risks regression analysis to yield more conservative confidence intervals. While competing-risks regression analyses do not have a goodness-of-fit statistic, we utilized the Cox model as a proxy. We confirmed non-violation of the proportional hazards assumption using “log-log” plots. A post-estimation function after the competing-risks regression model was utilized to generate cumulative incidence curves of recurrence, progression, and bladder cancer-related mortality rates. This was especially utilized to derive recurrence, progression, and mortality rates for each stage (Ta, Tis, and T1) after adjusting for covariates. We conducted all analyses with STATA software (College Station, Texas). All statistical tests were two-tailed, and the probability of a type I error was set at 0.05. The institutional review board at UCLA approved our study.
RESULTS
The vast majority of our cohort was younger than 80 years of age (61.3%), male (75.5%), white (91.0%), married (62.2%), without any comorbidity (65.7%), and diagnosed in the West (49.2%) with a poorly differentiated (78.1%) T1 tumor (57.5%) (Table 1). Of the 7,410 subjects, 3,828 (51.7%) died of all causes, while 981 (13.2%) died of bladder cancer (Figure 1). Moreover, 2,449 (33.1%) progressed and an additional 2,897 (39.1%) recurred without progression. Therefore 5,346 (72.2%) recurred, progressed or died of bladder cancer. Only 2,064 (27.8%) did not recur, progress, or die of bladder cancer.
Table 1.
Cohort characteristics (N=7410)
| Variables | Distribution | % |
|---|---|---|
| Age group | ||
| 66–69 | 985 | 13.3% |
| 70–74 | 1731 | 23.4% |
| 75–79 | 1825 | 24.6% |
| ≥80 | 2869 | 38.7% |
| Gender | ||
| Male | 5597 | 75.5% |
| Female | 1813 | 24.5% |
| Race/Ethnicity | ||
| White | 6742 | 91.0% |
| Black | 235 | 3.2% |
| Hispanic | 188 | 2.5% |
| Other | 245 | 3.3% |
| Marital status | ||
| Other | 2804 | 37.8% |
| Married | 4606 | 62.2% |
| Charlson Score | ||
| 0 | 4868 | 65.7% |
| 1 | 1635 | 22.1% |
| 2 | 587 | 7.9% |
| ≥3 | 320 | 4.3% |
| % with ≥4 years of college education | ||
| <15% | 1627 | 22.0% |
| 15–25% | 1906 | 25.7% |
| 25%–35% | 1608 | 21.7% |
| >35% | 2269 | 30.6% |
| Median household income | ||
| <$35,000 | 1303 | 17.6% |
| $35,000–$45,000 | 1770 | 23.9% |
| $45,000–$55,000 | 1893 | 25.5% |
| >$55,000 | 2444 | 33.0% |
| Region | ||
| West | 3643 | 49.2% |
| Midwest | 1541 | 20.8% |
| South | 706 | 9.5% |
| Northeast | 1520 | 20.5% |
| Institution Type | ||
| Non-academic non-cancer center | 5167 | 69.7% |
| Academic non-cancer center | 1614 | 21.8% |
| Academic cancer center | 165 | 2.2% |
| Unknown | 464 | 6.3% |
| Grade | ||
| Poorly differentiated | 5785 | 78.1% |
| Undifferentiated | 1625 | 21.9% |
| Stage | ||
| Ta | 2398 | 32.4% |
| Tis | 754 | 10.2% |
| T1 | 4258 | 57.5% |
Figure 1.
Schema of morbidity among our cohort of patients with high-grade non-muscle-invasive bladder cancer
Since competing causes of death may have precluded some patients (the elderly, the sick, or the poor) from developing recurrence of their disease, we sought to determine the cumulative incidence of recurrence, progression, or cause-specific mortality using multivariate competing-risks regression analysis. The incidence of recurrence, progression, and cause-specific mortality is depicted in Figure 2. The 2-, 5, and 10-year recurrence rates were 61.1%, 69.5%, and 74.3%. The 2-, 5-, and 10-year progression rates were 12.8%, 22.8%, and 33.3%. The 2-, 5, and 10-year cause-specific mortality rates were 6.5%, 10.2%, and 12.3%.
Figure 2.
Competing-risks regression analysis for recurrence, progression, and cause-specific mortality for bladder canser
On multivariate competing-risks regression analysis, only stage T1 was associated with a higher hazard of bladder cancer recurrence (HR 1.22; 95% CI 1.15–1.30) than those with Ta disease (Table 2). Patients with T1 disease had a 7% higher absolute risk of recurring at each time interval. Patient sociodemographics and socioeconomic status, stage Tis, and higher grade were not statistically significant predictors of recurrence.
Table 2.
Multivariate competing-risks regression analysis with 2-, 5-, and 10-year recurrence rates
| Hazard Ratio (95% CI) | 2-Yr Recurrence | 5-Yr Recurrence | 10-Yr Recurrence | p-value | |
|---|---|---|---|---|---|
| Overall | 61.1% | 69.5% | 74.3% | ||
| Age Group | |||||
| 66–69 | 1.00 (Referent) | 61.7% | 70.1% | 74.9% | … |
| 70–74 | 1.00 (0.91–1.10) | 61.7% | 70.1% | 74.9% | 0.98 |
| 75–79 | 1.03 (0.94–1.13) | 62.7% | 71.1% | 75.8% | 0.55 |
| ≥80 | 0.94 (0.86–1.03) | 59.4% | 67.8% | 72.7% | 0.16 |
| Gender | |||||
| Male | 1.00 (Referent) | 61.0% | 69.4% | 74.2% | … |
| Female | 1.01 (0.94–1.08) | 61.3% | 69.7% | 74.5% | 0.81 |
| Race/Ethnicity | |||||
| White | 1.00 (Referent) | 61.1% | 69.5% | 74.3% | … |
| Black | 1.05 (0.89–1.23) | 62.8% | 71.2% | 75.9% | 0.56 |
| Hispanic | 1.14 (0.97–1.34) | 66.0% | 74.2% | 78.8% | 0.11 |
| Other | 0.87 (0.74–1.02) | 55.9% | 64.3% | 69.2% | 0.08 |
| Marital Status | |||||
| Not Married | 1.00 (Referent) | 61.1% | 69.5% | 74.2% | … |
| Married | 1.00 (0.94–1.06) | 61.1% | 69.5% | 74.3% | 0.96 |
| Charlson Score | |||||
| 0 | 1.00 (Referent) | 60.8% | 69.2% | 74.0% | … |
| 1 | 1.06 (0.99–1.13) | 63.0% | 71.4% | 76.1% | 0.08 |
| 2 | 0.91 (0.81–1.01) | 57.3% | 65.7% | 70.6% | 0.08 |
| ≥3 | 1.03 (0.90–1.18) | 62.0% | 70.4% | 75.1% | 0.65 |
| % with Bachelor's Degree | |||||
| <15% | 1.00 (Referent) | 59.6% | 68.0% | 72.9% | … |
| 15–25% | 1.08 (0.99–1.17) | 62.4% | 70.8% | 75.5% | 0.07 |
| 25–35% | 1.05 (0.95–1.15) | 61.3% | 69.7% | 74.5% | 0.34 |
| >35% | 1.04 (0.94–1.14) | 60.9% | 69.3% | 74.1% | 0.49 |
| Median Household Income | |||||
| <$35,000 | 1.00 (Referent) | 62.1% | 70.5% | 75.2% | … |
| $35,000–$45,000 | 0.96 (0.87–1.05) | 60.6% | 69.0% | 73.8% | 0.38 |
| $45,001–$55,000 | 0.97 (0.87–1.06) | 60.8% | 69.2% | 74.0% | 0.48 |
| >$55,000 | 0.97 (0.88–1.08) | 61.1% | 69.5% | 74.3% | 0.62 |
| Grade | |||||
| Poorly Differentiated | 1.00 (Referent) | 61.0% | 69.4% | 74.2% | … |
| Undifferentiated | 1.01 (0.94–1.08) | 61.3% | 69.7% | 74.5% | 0.79 |
| Stage | |||||
| Ta | 1.00 (Referent) | 56.8% | 65.2% | 70.1% | … |
| Tis | 1.01 (0.92–1.12) | 57.3% | 65.7% | 70.6% | 0.79 |
| T1 | 1.22 (1.15–1.30) | 64.2% | 72.5% | 77.2% | <0.01 |
We found that advancing age (≥70) was associated with a statistically significant lower hazard of progression than those <70 years of age (Table 3). Patients 80 years of age or older had a lower absolute incidence of progression—4.5% lower at 2 years, 7.4% lower at 5 years, and 10.0% lower at 10 years—than patients younger than 70. Women (HR: 1.23; 95% CI 1.12–1.36) and black subjects (HR 1.37; 95% CI 1.11–1.68) had a higher hazard of progression than men and white subjects, respectively. At 2, 5, and 10 years, women had a 2.6%, 4.3%, and 5.8% higher incidence of progression than men. Black subjects had a 4.2%, 6.9%, and 9.2% higher incidence of progression than whites. Patients with undifferentiated grade (HR 1.24; 95% CI 1.13–1.19), stage Tis (HR 1.64; 95% CI 1.39–1.93) and T1 (HR 2.30; 95% CI 2.07–2.55) tumors had a higher hazard of progression than those with poorly differentiated and stage Ta tumors, respectively. Patients with T1 tumors had an absolute increase of 9.2%, 15.4%, and 21.0% in the incidence of progression at 2, 5, and 10 years after diagnosis than patients with Ta tumors.
Table 3.
Multivariate competing-risks regression analysis with 2-, 5-, and 10-year progression rates
| Hazard Ratio (95% CI) | 2-Yr Progression | 5-Yr Progression | 10-Yr Progression | p-value | |
|---|---|---|---|---|---|
| Overall | 12.8% | 22.8% | 33.3% | ||
| Age Group | |||||
| 66–69 | 1.00 (Referent) | 15.7% | 27.4% | 39.5% | … |
| 70–74 | 0.86 (0.76–0.98) | 13.7% | 24.2% | 35.1% | 0.02 |
| 75–79 | 0.84 (0.74–0.96) | 13.4% | 23.7% | 34.6% | 0.01 |
| ≥80 | 0.70 (0.61–0.79) | 11.2% | 20.0% | 29.5% | <0.01 |
| Gender | |||||
| Male | 1.00 (Referent) | 12.2% | 21.8% | 31.9% | … |
| Female | 1.23 (1.12–1.36) | 14.8% | 26.1% | 37.7% | <0.01 |
| Race/Ethnicity | |||||
| White | 1.00 (Referent) | 12.8% | 22.7% | 33.1% | … |
| Black | 1.37 (1.11–1.68) | 17.0% | 29.6% | 42.3% | <0.01 |
| Hispanic | 1.01 (0.78–1.31) | 12.9% | 22.9% | 33.5% | 0.92 |
| Other | 0.85 (0.68–1.07) | 11.0% | 19.7% | 29.1% | 0.17 |
| Marital Status | |||||
| Not Married | 1.00 (Referent) | 12.3% | 22.0% | 32.2% | … |
| Married | 1.07 (0.97–1.17) | 13.1% | 23.2% | 33.9% | 0.19 |
| Charlson Score | |||||
| 0 | 1.00 (Referent) | 13.1% | 23.2% | 33.9% | … |
| 1 | 0.97 (0.88–1.08) | 12.8% | 22.7% | 33.2% | 0.61 |
| 2 | 0.90 (0.76–1.06) | 11.8% | 21.1% | 31.0% | 0.20 |
| ≥3 | 0.81 (0.64–1.01) | 10.7% | 19.2% | 28.4% | 0.06 |
| % with Bachelor's Degree | |||||
| <15% | 1.00 (Referent) | 12.2% | 21.7% | 31.9% | … |
| 15–25% | 1.12 (0.98–1.27) | 13.6% | 24.0% | 34.9% | 0.08 |
| 25–35% | 1.11 (0.96–1.28) | 13.4% | 23.8% | 34.6% | 0.17 |
| >35% | 1.00 (0.86–1.18) | 12.2% | 21.8% | 32.0% | 0.96 |
| Median Household Income | |||||
| <$35,000 | 1.00 (Referent) | 12.9% | 22.9% | 33.5% | … |
| $35,000–$45,000 | 0.97 (0.84–1.12) | 12.6% | 22.4% | 32.8% | 0.72 |
| $45,001–$55,000 | 0.99 (0.96–1.28) | 12.8% | 22.6% | 33.1% | 0.86 |
| >$55,000 | 1.00 (0.85–1.19) | 13.0% | 23.0% | 33.6% | 0.95 |
| Grade | |||||
| Poorly Differentiated | 1.00 (Referent) | 12.3% | 21.8% | 32.0% | … |
| Undifferentiated | 1.24 (1.13–1.19) | 15.0% | 26.3% | 38.0% | <0.01 |
| Stage | |||||
| Ta | 1.00 (Referent) | 7.8% | 14.1% | 21.2% | … |
| Tis | 1.64 (1.39–1.93) | 12.4% | 22.0% | 32.3% | <0.01 |
| T1 | 2.30 (2.07–2.55) | 17.0% | 29.5% | 42.2% | <0.01 |
We found that advancing age (≥75) was associated with a statistically significant higher hazard of dying of bladder cancer than those <70 years of age (Table 4). Patients 80 years of age or older had a higher absolute incidence of dying of bladder cancer—3.5% at 2 years, 5.4% at 5 years, and 6.5% at 10 years—than patients younger than 70. Women (HR: 1.55; 95% CI 1.34–1.81), as well as black (HR 1.71; 95% CI 1.27–2.30) and Hispanic subjects (HR 1.46; 95% CI 1.03–2.09) had a higher hazard of dying of bladder cancer than men and whites, respectively. Patients with undifferentiated grade (HR 1.25; 95% CI 1.08–1.45), Tis (HR 2.11; 95% CI 1.57–2.78) and T1 (HR 3.06; 95% CI 2.54–3.69) tumors had a higher hazard of dying of bladder cancer than those with poorly differentiated and stage Ta tumors, respectively. Patients with T1 tumors had an absolute increase of 6.3%, 9.7%, and 11.6% in the incidence of dying of bladder cancer at 2, 5, and 10 years than patients with Ta tumors.
Table 4.
Multivariate competing-risks regression analysis with 2-, 5-, and 10-year cancer-specific mortality rates
| Hazard Ratio (95% CI) | 2-Yr Mortality | 5-Yr Mortality | 10-Yr Mortality | p-value | |
|---|---|---|---|---|---|
| Overall | 6.5% | 10.2% | 12.3% | ||
| Age Group | |||||
| 66–69 | 1.00 (Referent) | 4.9% | 7.8% | 9.4% | … |
| 70–74 | 1.03 (0.79–1.33) | 5.0% | 8.0% | 9.7% | 0.84 |
| 75–79 | 1.27 (1.00–1.62) | 6.2% | 9.8% | 11.9% | 0.05 |
| ≥80 | 1.75 (1.39–2.20) | 8.4% | 13.2% | 15.9% | <0.01 |
| Gender | |||||
| Male | 1.00 (Referent) | 5.8% | 9.2% | 11.1% | … |
| Female | 1.55 (1.34–1.81) | 8.9% | 13.9% | 16.8% | <0.01 |
| Race/Ethnicity | |||||
| White | 1.00 (Referent) | 6.4% | 10.0% | 12.2% | … |
| Black | 1.71 (1.27–2.30) | 10.6% | 16.5% | 19.9% | <0.01 |
| Hispanic | 1.46 (1.03–2.09) | 9.2% | 14.4% | 17.3% | 0.03 |
| Other | 0.72 (0.48–1.09) | 4.6% | 7.4% | 8.9% | 0.12 |
| Marital Status | |||||
| Not Married | 1.00 (Referent) | 6.7% | 10.6% | 12.8% | … |
| Married | 0.93 (0.81–1.08) | 6.3% | 10.0% | 12.0% | 0.37 |
| Charlson Score | |||||
| 0 | 1.00 (Referent) | 6.2% | 9.8% | 11.9% | … |
| 1 | 1.13 (0.96–1.33) | 7.0% | 11.0% | 13.3% | 0.14 |
| 2 | 1.08 (0.85–1.38) | 6.7% | 10.6% | 12.8% | 0.53 |
| ≥3 | 1.19 (0.87–1.61) | 7.3% | 11.5% | 13.9% | 0.27 |
| % with Bachelor's Degree | |||||
| <15% | 1.00 (Referent) | 5.9% | 9.3% | 11.2% | … |
| 15–25% | 1.10 (0.89–1.36) | 6.4% | 10.2% | 12.3% | 0.36 |
| 25–35% | 1.26 (1.01–1.58) | 7.3% | 11.6% | 14.0% | 0.04 |
| >35% | 1.08 (0.84–1.39) | 6.3% | 10.0% | 12.1% | 0.55 |
| Median Household Income | |||||
| <$35,000 | 1.00 (Referent) | 6.7% | 10.6% | 12.9% | … |
| $35,000–$45,000 | 1.00 (0.81–1.23) | 6.7% | 10.6% | 12.9% | 0.99 |
| $45,001–$55,000 | 0.98 (0.78–1.24) | 6.6% | 10.5% | 12.7% | 0.88 |
| >$55,000 | 0.88 (0.68–1.14) | 6.0% | 9.5% | 11.5% | 0.36 |
| Grade | |||||
| Poorly Differentiated | 1.00 (Referent) | 6.1% | 9.7% | 11.8% | … |
| Undifferentiated | 1.25 (1.08–1.45) | 7.7% | 12.1% | 14.6% | <0.01 |
| Stage | |||||
| Ta | 1.00 (Referent) | 3.2% | 5.1% | 6.2% | … |
| Tis | 2.11 (1.57–2.78) | 6.6% | 10.4% | 12.6% | <0.01 |
| T1 | 3.06 (2.54–3.69) | 9.5% | 14.8% | 17.8% | <0.01 |
DISCUSSION
Our study has three principal findings. First, while we are not the first to report on the recurrence and progression rates in patients with non-muscle-invasive bladder cancer, this is the first study to examine the natural history of the disease from a population standpoint. We found that nearly three-fourths of patients with high-grade non-muscle-invasive bladder cancer will develop a recurrence at 10 years—41% will recur without progression, while an additional 33% will progress to myoinvasive disease. Furthermore, among those who progress, 40% (981 out of 2449) will die of bladder cancer. The dismal prognosis of non-muscle-invasive bladder cancer progressing to myoinvasive disease highlights the need to intensely survey this high-risk population early on. However, in a previous analysis of 4,545 patients with high-grade non-muscle-invasive bladder cancer, we discovered that 62.5% of patients did not undergo at least 1 cystoscopy, 1 cytology, and 1 instillation of immunotherapy during the initial two years after diagnosis.7
Our second important finding is that while gender and race do not correlate with recurrence rates, they are significantly associated with progression and cause-specific mortality rates. Women and black subjects had a higher likelihood of progressing and ultimately dying of bladder cancer than did men and whites, respectively. While Hispanic subjects had a higher risk of dying of bladder cancer, they did not have a corresponding increase in aggressive treatment. Sociodemographic variation in aggressive treatment may in part be attributed to the growing misutilization of immunotherapy among patients who frequently recur or ultimately progress to myoinvasive disease.14, 15 Additionally, advancing age was associated with a higher rate of bladder cancer mortality. The higher cancer-specific mortality rate among the elderly may be a result of either quality of care or response (or lack thereof) to our current treatments. To the former, we have previously demonstrated that the elderly (those 80 years of age and older) were less likely to undergo the recommended number of surveillance (cytology and imaging) and treatment (immunotherapy) measures.7 While provider-level variation is a significant predictor of non-compliance for some measures (e.g. cytology and immunotherapy), we cannot ignore patient-level factors in this at-risk population—non-compliance may be due to their poor performance status, mobility, lack of social support, and access to transportation. Asking octogenarians to come in for weekly intravesical instillations of immunotherapy may be futile if arrangements are not made ahead of time. As to the latter, Joudi et al has demonstrated a lower response rate to immunotherapy and worsened recurrence-free survival among the elderly.16 Hence, we may need to incorporate multi-agent and sequential intravesical chemotherapy for the elderly.17 It is therefore not surprising that we found the elderly progressing to invasive disease, but due to their age and health they are not offered aggressive treatment (e.g. cystectomy, radiotherapy, or systemic chemotherapy), and hence have a higher rate of bladder cancer-related mortality. Some may contend that even among the elderly, aggressive treatment for invasive disease is associated with improved survival.18, 19 However, there is a dearth of data quantifying the potential benefit of aggressive treatment such as a radical cystectomy among the elderly for recurrent non-muscle-invasive bladder cancer. In other words, should we expose all patients with recurrent non-muscle-invasive bladder cancer—even the elderly with significant comorbidity—to aggressive treatment?
Third, the integration of treatment codes in a competing-risks regression analysis serves as an ideal model for future studies to quantify recurrence and progression rates of bladder cancer. Our a priori definitions of recurrence and progression are commensurate with other single institution series describing 48%–80% recurrence and 18%–53% progression rates.20–28 In fact our 5-year recurrence (69%), progression (23%), and mortality (10%) rates for high-grade cancers are very similar to the largest single institution series (1529 patients) that reported recurrence, progression, and mortality rates of 61%, 19%, and 10%, respectively.28
While our sample size is large, our study has methodological limitations. As with any observational study, omitted-variable bias may impact aggressive treatment rates, and thus the imputed progression rates. Some may contend that our five-year recurrence and progression rates are significantly higher than the 48% recurrence and 11% progression long-term rates reported from a combined analysis of seven European Organization for Research and Treatment Cancer (EORTC) trials.29 We attribute this to differences in cohort characteristics (all grades vs high grade). If one examines the two highest risk groups in the EORTC trial, their averaged five-recurrence and progression rates are 70% and 31%, similar to the 69% and 23% in our study. Our lower five-year progression rate may in part be attributed to inherent differences in the way European participants enrolled in clinical trials are treated when compared with a SEER-Medicare population—78% of the European patients received early adjuvant intravesical immunotherapy, while only 30% of our cohort received at least one instillation. Furthermore, our definition of progression hinges on treatment codes (radical cystectomy, radiotherapy and systemic chemotherapy) and it is not unreasonable to deduce that the 8% difference in five-year progression rates may in part be attributed to level of under-treatment in the SEER-Medicare population as reported by others.2, 3 Despite the lack of granularity in measuring recurrence and progression rates, our estimates are well within the range of other studies. A second limitation was the utility of treatment codes to define progression instead of stage-based progression. For instance, a patient that progresses from a unifocal Ta tumor to multifocal T1 disease that was subsequently treated with immunotherapy does not meet the definition of progression. Similarly a patient with T1 disease that fails immunotherapy and ultimately develops myoinvasive disease treated with radical transurethral resection does not meet the definition of progression. This incongruity is especially evident among those 80 years of age and older—who had a lower incidence of progression yet a higher incidence of bladder cancer death. This is more clearly seen with the derived mortality-to-progression ratio of patients: This ratio increases from 0.24 in the 66–69 age group, to 0.28 in the 70–74 group to 0.34 in the 75–79 group, and to 0.54 among those ≥80. This suggests that the elderly are likely being undertreated for tumor progression, hence the lower claims-based definition of “progression”. All the while, the risk of dying of bladder cancer increases with advancing age. In an ideal setting, we would be able to conduct this with pathologic data to confirm true tumor progression. However, SEER and the local cancer registries do not capture or record pathologic data beyond 6 months of diagnosis. Therefore, we must rely on the behavior of high-grade bladder cancer (to mandate aggressive treatment or cancer-related mortality) and claims data to quantify recurrence and progression. While this is a limitation that we acknowledge, it does not detract from the significance of our findings for two reasons. One is that if elderly patients were truly progressing to myoinvasive or extravesical disease, this would mean that they were undergoing endoscopic resections at a higher rate, which they are not. Additionally, true measure of tumor progression (myoinvasive disease and extravesical extension) may not be reliably captured with endoscopic resections. Instead, we would have to examine the cystectomy specimen or rely on imaging for clinical progression to ascertain pathological progression. If a patient undergoes a cystectomy (or any aggressive treatment), then we would capture that progression. If they don't, we can rely on the nature of untreated high-grade disease, which if left untreated will likely result in bladder cancer death. And if a patient dies of another cause, our competing-risks regression model accounts for competing-risks of death. Another limitation is that the findings of the current study may not be generalizable to those patients who are aged <65 years or who have alternative forms of insurance coverage. However, 75% of all patients with bladder cancer are aged ≥65 years, and the vast majority of the elderly have Medicare benefits.30, 31 Last, our definition of recurrence (endoscopic resection or biopsy) may not be entirely accurate. Some endoscopic procedures may be performed for suspicious lesions, which turn out to be benign. Alternatively, some patients may have had small asymptomatic papillary lesions that are not treated. However, we anticipate that the majority of patients with high-grade cancer who develop a recurrence will undoubtedly be treated or ultimately progress and die of their disease (if left untreated).
Despite these limitations, our findings serve to alert patients and providers to the significant morbidity of high-grade bladder cancer. Also, when these lesions progress to invasive disease, patients are at significant risk of bladder-cancer related death. It is for this reason that use of intravesical therapy—to minimize recurrences and progression to invasive disease—in combination with an intense surveillance schedule are outlined in the AUA and NCCN guidelines. However, the expectation that practice patterns will change overnight because of our findings or publication of an update from the AUA or NCCN is naïve. Overcoming non-compliance with guideline-recommended care is likely to be a challenge for a number of reasons—some modifiable and some not. As to the former, we believe that quality-improvement initiatives, like the Urological Surgical Quality Collaborative or the Surgical Clinical Outcomes Assessment Program developed by researchers at the University of Michigan and Washington, respectively, would be needed to see substantive change in practice patterns.32–34 These initiatives were developed with the understanding that in order to improve quality, health services researchers must first qualitatively describe logistical factors that limit access to evidence-based care. Additionally, community practices, where most urological care is provided, may not be conducive to the most aggressive treatment options; rather, patients may benefit from referral to academic settings.35 And while regionalization to high-volume settings may improve quality of care, they do come at the expense of impeding access and timely treatment.
CONCLUSIONS
Even though most patients with high-grade, non-muscle-invasive bladder cancer do not die of their disease, the vast majority endures the morbidity of recurrence and progression their disease. Increasing efforts should be made to offer patients intravesical therapy with the goal of minimizing the incidence of recurrences. Furthermore, the high recurrence rate during the first two years of diagnosis warrants an intense surveillance schedule.
Acknowledgments
FUNDING This work was supported by the American Cancer Society (117496-PF-09-147-01-CPHPS (Principal Investigator: KC)); Ruth L. Kirschstein National Research Service Award Extramural (1 F32 CA144461-01 (Principal Investigator: KC)); Jonsson Comprehensive Cancer Center Seed Grant (Principal Investigator: KC); National Institutes of Health Loan Repayment Program (Principal Investigator: KC); and National Institute of Diabetes and Digestive and Kidney Diseases (N01-DK-1-2460 (Principal Investigator: MSL))
Footnotes
Disclosures: No financial disclosures to report
REFERENCES
- 1.Cancer Statistics 1984–2011 American Cancer Society Surveillance Research [Google Scholar]
- 2.Fedeli U, Fedewa SA, Ward EM. Treatment of muscle invasive bladder cancer: evidence from the National Cancer Database, 2003 to 2007. The Journal of urology. 2011;185(1):72–8. doi: 10.1016/j.juro.2010.09.015. [DOI] [PubMed] [Google Scholar]
- 3.Gore JL, Lai J, Setodji CM, Litwin MS, Saigal CS. Mortality increases when radical cystectomy is delayed more than 12 weeks: results from a Surveillance, Epidemiology, and End Results-Medicare analysis. Cancer. 2009;115(5):988–96. doi: 10.1002/cncr.24052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Herr HW, Dotan Z, Donat SM, Bajorin DF. Defining optimal therapy for muscle invasive bladder cancer. J Urol. 2007;177(2):437–43. doi: 10.1016/j.juro.2006.09.027. [DOI] [PubMed] [Google Scholar]
- 5.Madeb R, Golijanin D, Noyes K, et al. Treatment of nonmuscle invading bladder cancer: do physicians in the United States practice evidence based medicine? The use and economic implications of intravesical chemotherapy after transurethral resection of bladder tumors. Cancer. 2009;115(12):2660–70. doi: 10.1002/cncr.24311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Huang GJ, Hamilton AS, Lo M, Stein JP, Penson DF. Predictors of intravesical therapy for nonmuscle invasive bladder cancer: results from the surveillance, epidemiology and end results program 2003 patterns of care project. J Urol. 2008;180(2):520–4. doi: 10.1016/j.juro.2008.04.016. discussion 24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Chamie K, Saigal CS, Lai J, et al. Compliance with guidelines for patients with bladder cancer: variation in the delivery of care. Cancer. 2011;117(23):5392–401. doi: 10.1002/cncr.26198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chamie K, Saigal CS, Lai J, et al. Quality of care in patients with bladder cancer: A case report? Cancer. 2011 doi: 10.1002/cncr.26402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Warren JL, Klabunde CN, Schrag D, Bach PB, Riley GF. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Med Care. 2002;40(8 Suppl):IV-3–18. doi: 10.1097/01.MLR.0000020942.47004.03. [DOI] [PubMed] [Google Scholar]
- 10.Bach PB, Guadagnoli E, Schrag D, Schussler N, Warren JL. Patient demographic and socioeconomic characteristics in the SEER-Medicare database applications and limitations. Med Care. 2002;40(8 Suppl):IV-19–25. doi: 10.1097/00005650-200208001-00003. [DOI] [PubMed] [Google Scholar]
- 11.Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. J Clin Epidemiol. 2000;53(12):1258–67. doi: 10.1016/s0895-4356(00)00256-0. [DOI] [PubMed] [Google Scholar]
- 12.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–83. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 13.Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. Journal of the American Statistical Association. 1999;94(446):496–509. [Google Scholar]
- 14.Lambert EH, Pierorazio PM, Olsson CA, Benson MC, McKiernan JM, Poon S. The increasing use of intravesical therapies for stage T1 bladder cancer coincides with decreasing survival after cystectomy. BJU Int. 2007;100(1):33–6. doi: 10.1111/j.1464-410X.2007.06912.x. [DOI] [PubMed] [Google Scholar]
- 15.Morris DS, Weizer AZ, Ye Z, Dunn RL, Montie JE, Hollenbeck BK. Understanding bladder cancer death: tumor biology versus physician practice. Cancer. 2009;115(5):1011–20. doi: 10.1002/cncr.24136. [DOI] [PubMed] [Google Scholar]
- 16.Joudi FN, Smith BJ, O'Donnell MA, Konety BR. The impact of age on the response of patients with superficial bladder cancer to intravesical immunotherapy. The Journal of Urology. 2006;175(5):1634–9. doi: 10.1016/S0022-5347(05)00973-0. discussion 39–40. [DOI] [PubMed] [Google Scholar]
- 17.Breyer BN, Whitson JM, Carroll PR, Konety BR. Sequential intravesical gemcitabine and mitomycin C chemotherapy regimen in patients with non-muscle invasive bladder cancer. Urologic oncology. 2010;28(5):510–4. doi: 10.1016/j.urolonc.2008.11.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chamie K, Hu B, Devere White RW, Ellison LM. Cystectomy in the elderly: does the survival benefit in younger patients translate to the octogenarians? BJU Int. 2008 doi: 10.1111/j.1464-410X.2008.07636.x. [DOI] [PubMed] [Google Scholar]
- 19.Hollenbeck BK, Miller DC, Taub D, et al. Aggressive treatment for bladder cancer is associated with improved overall survival among patients 80 years old or older. Urology. 2004;64(2):292–7. doi: 10.1016/j.urology.2004.03.034. [DOI] [PubMed] [Google Scholar]
- 20.Heney NM, Ahmed S, Flanagan MJ, et al. Superficial bladder cancer: progression and recurrence. The Journal of urology. 1983;130(6):1083–6. doi: 10.1016/s0022-5347(17)51695-x. [DOI] [PubMed] [Google Scholar]
- 21.Jakse G, Loidl W, Seeber G, Hofstadter F. Stage T1, grade 3 transitional cell carcinoma of the bladder: an unfavorable tumor? The Journal of urology. 1987;137(1):39–43. doi: 10.1016/s0022-5347(17)43864-x. [DOI] [PubMed] [Google Scholar]
- 22.Abel PD, Hall RR, Williams G. Should pT1 transitional cell cancers of the bladder still be classified as superficial? British journal of urology. 1988;62(3):235–9. doi: 10.1111/j.1464-410x.1988.tb04325.x. [DOI] [PubMed] [Google Scholar]
- 23.Pauwels RP, Schapers RF, Smeets AW, Debruyne FM, Geraedts JP. Grading in superficial bladder cancer. (1). Morphological criteria. British journal of urology. 1988;61(2):129–34. doi: 10.1111/j.1464-410x.1988.tb05060.x. [DOI] [PubMed] [Google Scholar]
- 24.Sarkis AS, Dalbagni G, Cordon-Cardo C, et al. Nuclear overexpression of p53 protein in transitional cell bladder carcinoma: a marker for disease progression. Journal of the National Cancer Institute. 1993;85(1):53–9. doi: 10.1093/jnci/85.1.53. [DOI] [PubMed] [Google Scholar]
- 25.Holmang S, Hedelin H, Anderstrom C, Johansson SL. The relationship among multiple recurrences, progression and prognosis of patients with stages Ta and T1 transitional cell cancer of the bladder followed for at least 20 years. The Journal of urology. 1995;153(6):1823–6. discussion 26–7. [PubMed] [Google Scholar]
- 26.Herr HW. Tumour progression and survival in patients with T1G3 bladder tumours: 15-year outcome. Br J Urol. 1997;80(5):762–5. doi: 10.1046/j.1464-410x.1997.00431.x. [DOI] [PubMed] [Google Scholar]
- 27.Holmang S, Hedelin H, Anderstrom C, Holmberg E, Johansson SL. The importance of the depth of invasion in stage T1 bladder carcinoma: a prospective cohort study. The Journal of urology. 1997;157(3):800–3. doi: 10.1016/s0022-5347(01)65044-4. discussion 04. [DOI] [PubMed] [Google Scholar]
- 28.Millan-Rodriguez F, Chechile-Toniolo G, Salvador-Bayarri J, Palou J, Algaba F, Vicente-Rodriguez J. Primary superficial bladder cancer risk groups according to progression, mortality and recurrence. The Journal of Urology. 2000;164(3 Pt 1):680–4. doi: 10.1016/s0022-5347(05)67280-1. [DOI] [PubMed] [Google Scholar]
- 29.Sylvester RJ, van der Meijden AP, Oosterlinck W, et al. Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials. European urology. 2006;49(3):466–5. doi: 10.1016/j.eururo.2005.12.031. discussion 75–7. [DOI] [PubMed] [Google Scholar]
- 30.Ries LAGEM, Kosary CL, Hankey BF, Miller BA, Clegg L, Mariotto A, Feuer EJ, Edwards BK. SEER Cancer Statistics Review, 1975–2002. National Cancer Institute; Bethesda, MD: http://seer.cancer.gov/csr/1975_2002/, based on November 2004 SEER data submission, posted to the SEER web site 2005 2005. [Google Scholar]
- 31.Federal Interagency Forum on Aging-Related Statistics . Older Americans update 2006: Key indicators of well-being. U.S. Government Printing Office; Washington, DC: Jul, 2006. [Google Scholar]
- 32.Miller DC, Murtagh DS, Suh RS, Knapp PM, Dunn RL, Montie JE. Establishment of a urological surgery quality collaborative. The Journal of urology. 2010;184(6):2485–90. doi: 10.1016/j.juro.2010.08.015. [DOI] [PubMed] [Google Scholar]
- 33.Miller DC, Murtagh DS, Suh RS, et al. Regional collaboration to improve radiographic staging practices among men with early stage prostate cancer. The Journal of Urology. 2011;186(3):844–9. doi: 10.1016/j.juro.2011.04.078. [DOI] [PubMed] [Google Scholar]
- 34.Burks FN, Liu AB, Suh RS, et al. Understanding the use of immediate intravesical chemotherapy for patients with bladder cancer. The Journal of Urology. 2012;188(6):2108–13. doi: 10.1016/j.juro.2012.08.044. [DOI] [PubMed] [Google Scholar]
- 35.Hollenbeck BK, Taub DA, Miller DC, Dunn RL, Montie JE, Wei JT. The regionalization of radical cystectomy to specific medical centers. J Urol. 2005;174(4 Pt 1):1385–9. doi: 10.1097/01.ju.0000173632.58991.a7. discussion 89. [DOI] [PubMed] [Google Scholar]