- Email: [email protected]
National treatment trends among older patients with T1-localized renal cell carcinoma1
Urologic Oncology: Seminars and Original Investigations ] (2016) ∎∎∎–∎∎∎
Original article
National treatment trends among older patients with T1-localized renal cell carcinoma1 Simon P. Kim, M.D., M.P.H.a,b,c,*, Cary P. Gross, M.D.c,d, Neal Meropol, M.D.b,e, Alexander Kutikov, M.D.f, Marc C. Smaldone, M.D.f, Nilay D. Shah, Ph.D.g, James B. Yu, M.D.h, Sarah Psutka, M.D., M.S.i, Jonathon Kiechle, M.D.a, Robert Abouassaly, M.D., M.S.a a
Urology Institute, Center of Outcomes and Health Care Quality, University Hospitals Cleveland Medical Center, Cleveland, OH b Seidman Cancer Center, University Hospitals Cleveland Medical Center, Cleveland, OH c Cancer Outcomes Public Policy and Effectiveness Research (COPPER), Yale University, New Haven, CT d Department of Medicine, Yale University, New Haven, CT e Case Comprehensive Cancer Center, Case Western Reserve University, Cleveland, OH f Division of Urologic Oncology, Philadelphia, PA g Division of Health Care Policy and Research, Mayo Clinic, Rochester, MN h Department of Radiation Oncology, Yale University, New Haven, CT i Division of Urologic Surgery, Cook County Health and Hospitals Systems, Chicago, IL Received 3 July 2016; received in revised form 30 September 2016; accepted 7 October 2016
Abstract Objective: To assess the national trends in treatment of localized renal tumors among older patients with limited life expectancy. Materials and methods: Using the National Cancer Database, we identified older patients (Z70 y) diagnosed with T1 renal cell carcinoma from 2002 to 2011. Primary outcome was the initial treatment—partial nephrectomy (PN), radical nephrectomy, EM, and ablation. Multivariable logistic regression analysis stratified by tumor size (o2, 2–3.9, or 4–7 cm) and age groups (70–79 and Z80 y) was used to identify covariates associated with different treatments. Results: Among 41,518 older patients with T1 renal cell carcinoma renal tumors, most were treated with radical nephrectomy (59.0%) followed by PN (20.0%) and ablation (8.4%). Only 12.6% were managed by EM. Among older patients aged 70 to 79 years with renal tumors 2 to 3.9 cm, PN was used more frequently in 2008 to 2009 (odds ratio [OR] ¼ 1.32; P ¼ 0.001) and 2010 to 2011 (OR ¼ 1.87; P o 0.001) compared to 2002 to 2003 and at academic hospitals (OR ¼ 1.91; P o 0.001) compared to community hospitals. Similar trends were observed for patients aged 70 to 79 years with 4 to 7 cm tumors and for patients aged Z80 years across renal tumor sizes. Conclusions: Among older patients with localized renal tumors and limited life expectancy, most are treated surgically with a growing use of PN. A smaller proportion of older patients are managed by EM in the United States. r 2016 Elsevier Inc. All rights reserved.
Keywords: Advanced age; Kidney cancer; Nephrectomy; Renal cell carcinoma; Treatment decisions
1. Introduction Two population trends are occurring in the United States that would increase complexity of treatment decisions for small 1 Dr. Simon P. Kim is supported by a career development award from the Conquer Cancer Foundation from the American Society of Clinical Oncology. * Corresponding author. Fax: þ1-216-844-1100. E-mail address: [email protected] (S.P. Kim).
http://dx.doi.org/10.1016/j.urolonc.2016.10.008 1078-1439/r 2016 Elsevier Inc. All rights reserved.
renal masses (SRMs). The incidence of SRMs has been gradually rising such that clinical T1 renal tumors represent most incident cases, which has been attributable to the growing use of imaging [1,2]. Approximately 60,000 patients would face a diagnosis of renal cell carcinoma (RCC) making it the sixth most common cancer in the United States this year [3]. Against this backdrop, the U.S. population is growing older. The average life expectancies for both men and women have been steadily increasing such that a fifth of the U.S. population would be aged 65 years or older by 2030 [4]. As a result, the rising
2
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
proportion of elderly patients with incidentally detected localized renal tumors with uncertain malignant potential represents a management challenge with significant health policy implications. Clinical practice guidelines currently endorse partial nephrectomy (PN) for SRMs amenable to surgical resection [5–7]. Other possible treatment options for SRMs include radical nephrectomy (RN), ablation, or expectant management (EM). Although PN has become more technically feasible and safer with minimally invasive surgery, such as robotic PN, it still carries some morbidity and bleeding risks that may be poorly tolerated among older patients [8,9]. With the changes in clinical practice guidelines and rapid dissemination of robotic surgery, the national rates of PN have been gradually rising in the United States [10]. However, EM has become an increasingly accepted disease management strategy for SRMs based on the growing number of studies suggesting the modest annual growth rates of the renal tumor and low risk of metastatic potential and cancer-related mortality, in particular among patients with a limited life expectancy [11–13] Yet, an important knowledge gap is the contemporary trends in the treatment of SRMs among older patients in the United States, as some older patients may be exposed to aggressive treatments and its associated morbidities without receiving a survival benefit. In this context, we assessed the contemporary national treatment trends of older patients (470 y) diagnosed with SRMs. 2. Material and methods 2.1. Data source We queried the National Cancer Database (NCDB), a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society, to examine surgical treatment in elderly patients with SRMs [14]. The NCDB is a national oncology dataset obtained from more than 1,500 hospitals that contains no patient- or physician-identifying information. Approximately 70% of newly diagnosed malignancies are captured in the NCDB. 2.2. Study population We identified all adult patients aged Z70 with kidney masses identified as r7 cm or stage T1a or T1b from the years 2002 to 2011. Patients were only included if RCC was their first and solitary cancer diagnosis. Exclusion criteria also included the presence of clinical nodal (N1–3) or metastatic disease (M1), age less than 70 years, and primary tumor in the renal pelvis or ureter. 2.3. Covariates and outcomes Age at diagnosis, race, sex, 2000 census tract annual median income, insurance status (private, Medicare, Medicaid, and
other), geographic region (East, Midwest, West, and South), location (urban, rural, and metro), Charlson-Deyo comorbidity score, education status, and year of treatment were assigned for each patient using NCDB data [15]. Hospital types were stratified using the standard NCDB codes for hospitals. Per NCDB guidelines, academic/research hospitals must be primarily affiliated with a medical school or be a National Cancer Institute-designated comprehensive cancer center. The distinction between comprehensive community and community cancer programs is based on overall case volume and the number of ancillary staff available to help treat oncology patients [16]. The NCDB classifies approximately 20% of institutions as academic/research hospitals, 39% as comprehensive community cancer centers, 35% as community cancer centers, and 6% as other cancer centers [17]. Treatment modalities were defined using NCDB codes for surgical treatment of the primary site to determine receipt of PN, RN, local ablative therapy, or EM. We classified patients who did not receive active treatment as undergoing expectant management. 2.4. Statistical analysis The primary outcome was the initial treatment administered—EM, PN, RN, and ablation. Bivariate associations between treatment type (EM, local ablation, or surgery) and patient characteristics were analyzed using Pearson’s chisquare test. Multivariable logistic regression analysis stratified by tumor size (o2 cm, 2–3.9 cm, or 4–7 cm) and age groups (70–79 and Z80 y) was used to identify covariates associated with different treatments adjusting for patient demographic and clinical characteristics. Stata MP version 11.2 was used to perform all statistical analyses, and a 2-sided P o 0.05 was used to determine statistical significance [18]. 3. Results From 2002 to 2011, we identified 41,518 patients diagnosed with RCC in the NCDB. As shown in Table 1, most patients were aged between 70 and 79 years (71.4%), white (77.3%), and relatively healthy with no Charlson comorbidities (63.0%). Most patients were treated in urban locations (80.8%) and at comprehensive community hospitals (55.3%), whereas only a third of patients were treated at academic hospitals (32.6%). During the study interval, most patients had a clear cell RCC histology (83.0%) with most patients receiving some form of treatment with RN (59.0%), PN (20.0%), and ablation (8.4%). Only 12.6% of patients were managed with EM. Significant temporal trends in the national utilization of different treatments and EM were found from 2002 to 2011 (Fig.). From 2002–2003 to 2010–2011, there was marked decrease in the proportion of patients undergoing RN from 73.5% to 47.3%, respectively, whereas the national utilization of PN rose from 15.2% to 27.4%, respectively (both
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7 Table 1 Clinical and demographic covariates (n ¼ 41,518)
Table 1 Continued
Variable
% (n)
Age (y) 70–79 Z80
71.4 (28,642) 28.6 (11,876)
Female
46.6 (19,363)
Race White Black Hispanic Other
77.3 8.1 11.0 3.6
Charlson comorbidity index 0 1 Z2
63.0 (24,145) 26.2 (10,023) 10.8 (4,132)
Location Urban Metro Rural
80.7 (31,565) 16.8 (6,574) 2.5 (957)
Median zip code income o30,000 30,000–34,999 35,000–45,999 Z46,000
14.1 18.8 29.0 38.1
(5,559) (7,373) (11,399) (14,990)
Health insurance Private Medicaid Medicare Other
10.6 1.1 85.6 2.7
(10.6) (457) (35,552) (1,103)
Geographic region Northeast South Midwest West Hospital type Community Comprehensive community Academic Other
3
(32,101) (3,377) (4,556) (1,484)
21.1 37.4 26.2 15.3 9.7 55.3 32.6 2.4
(4,027) (22,947) (13,530) (1,104)
Location Urban Metro Rural
80.8 (31,565) 16.8 (6,574) 2.4 (957)
Tumor histology Clear cell Papillary Chromophobe Sarcomatoid Other
81.7 10.7 4.7 0.6 2.2
Primary tumor size (cm) o2 2–3.9 Z4
9.6 (3,975) 49.3 (20,472) 41.1 (17,071)
(33,923) (4,469) (1,978) (247) (901)
Variable
% (n)
Primary treatment PN RN Ablation EM
20.0 59.0 8.4 12.6
(8,310) (24,509) (3,458) (5,241)
Time interval 2002–2003 2004–2005 2006–2007 2008–0AS 2010–2011
15.7 18.6 21.5 22.1 22.1
(6,515) (7,708) (8,919) (9,179) (9,197)
P o 0.001 for trend). However, use of renal mass ablation rose from 1.6% in 2002–2003 to 11.7% in 2010–2011. Similarly, the percentage of patients with clinical T1 RCC managed with EM gradually rose during the study interval from 9.8% in 2002–2003 to 13.6% in 2010–2011 (P o 0.001 for trend). On multivariable analysis, our results indicate that the year of treatment and the type of hospital were associated with the primary outcomes when stratified by tumor size and age groups. Among patients aged 70 to 79 years (Table 2), patients had higher odds ratios (OR) for receipt of PN in 2008 to 2009 and 2010 to 2011 for SRMs measuring 2 to 3.9 cm and 4 to 7 cm, respectively, compared with 2002 to 2003 (both P o 0.01). When treated at academic hospitals, patients were also more likely to undergo PN in ORs compared to community hospitals across all different tumor sizes in this age subgroup (all P o 0.05). For instance, patients with T1b renal tumors (4–7 cm) had an OR of 2.74 of undergoing PN at academic hospitals (95% CI: 2.10–3.60). Likewise, RN was associated with decreased use starting in 2004–2005 to 2010–2011 compared to 2002–2003 and when treated at academic hospitals (all P o 0.001). Both ablation and EM also had higher ORs over time and when treated at academic hospitals for larger tumors that were 2.0 to 3.9 and 4 to 7 cm as well (all P o 0.05). Among the subgroup of patients aged Z80 years, the multivariable logistic regression analysis also demonstrated different treatments and EM associated with time intervals and hospital characteristics (Table 3). When compared with 2002 to 2003, PN was also associated with higher OR for renal tumors in 2010 to 2011 measuring o2 cm (OR ¼ 1.44, 95% CI: 1.03–2.01), 2 to 3.9 cm (OR ¼ 1.48, 95% CI: 1.08–2.06), and 4 to 7 cm (OR ¼ 2.04, 95% CI: 1.20–3.45). Likewise, patients who were Z80 years were more likely to undergo PN at academic hospitals across all different tumor size groups compared to community hospitals (all P o 0.05). Older patients were also less likely to undergo RN over more recent years starting from 2006–2007 to 2010–2011 (P o 0.05). Older patients
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
4 80.0% 70.0% 60.0% 50.0%
PN RN Ablation EM
40.0% 30.0% 20.0% 10.0% 0.0%
2002-03
2004-05
2006-07
2008-09
2010-11
Fig. Temporal trends of different treatments and EM over time. All treatments and EM with P o 0.001 for trend.
in this subgroup also had higher OR of undergoing renal mass ablations, however, when treated in more recent years for all time intervals across all tumor sizes (all P o 0.05). Although EM was used more frequently over time with higher ORs from 2006–2007 to 2010–2011 compared to 2002–2003 (all P o 0.05), academic hospitals were associated with statistically significant lower ORs compared to community hospitals for renal tumors sized 2 to 3.9 cm (OR ¼ 0.84, 95% CI: 0.50–0.79) and 4 to 7 cm (OR ¼ 0.65, 95% CI: 0.50–0.86). Likewise, patients who were Z80 years and treated at comprehensive community hospitals were also less likely to undergo EM for renal tumors 2 to 3.9 cm and 4 to 7 cm when compared to community hospitals (both P o 0.05).
4. Discussion Our study presents important information at a time when the patient population is growing older, and there is a rising incidence of SRMs in the United States, along with a growing trend in the use of surgical therapy, in particular, PN [10,19]. Yet, it is essential to recognize that clinical practice guidelines do not identify specific patient characteristics where PN is most efficacious for improving outcomes other than tumor location and complexity. Against this backdrop, our study has several key findings. First, many older patients in the NCDB cohort with clinical T1 RCC received some form of primary therapy (87.4%), where approximately 80% underwent PN or RN. Furthermore, the use of PN nearly doubled with a concomitant drop by a third of RN from 2002 to 2011 among patients aged 470 years. Although previous studies have documented the gradual increases in PN for all patients, we found that older patients are undergoing PN, especially considering the risk of complications with bleeding compared to conventional RN or EM [20]. Indeed, the comparative effectiveness research about whether this segment of the patient population benefits from PN
compared to RN is mixed. Recently, Tan et al. [21] performed an instrumental variable analysis to assess the differences in overall survival and cancer-specific survival among Medicare beneficiaries treated with PN or RN from 1992 to 2007. Although this population-based cohort study demonstrated a lower hazard ratio for all-cause mortality with PN compared to RN (hazard ratio ¼ 0.54, 95% CI: 0.34–0.85) for all Medicare beneficiaries (age 4 65 y), PN did not confer any statistically significant difference in all-cause mortality in subgroup of analysis restricted to patients Z75 years old. In addition, Kutikov et al. [22] developed a nomogram for patients with T1a renal tumors to predict 5- and 10-year noncancer related, kidney cancer-related, and other cancer-related mortality using surveillance, epidemiology, and end results (SEER) data from 1988 to 2003 in order to objectively inform patients about the benefits of treatments. For example, a 75-year-old patient would more likely have a greater risk of non–kidney cancer-related death than RCC mortality at 5-years based on the nomogram (19% vs. 5%). Similarly, Sun et al. [23] also performed a competing risk analysis of Medicare beneficiaries comparing surgical management (PN or RN) vs. nonsurgical management from SEER-Medicare between 1998 and 2005, where patients Z75 years old had negligible survival benefit compared to EM. Taken together, our study suggests that older patients with limited life expectancy are more frequently undergoing PN with a paucity of evidence to support such aggressive treatment and better outcomes. In an era of individualized medicine, shared decision-making and clinical implementation of decision models would facilitate better patient understanding and knowledge about the risks and benefits of all treatment options allowing them to make an informed decision. Second, our study clearly demonstrates that EM is used in most older patients with T1 RCC, though its trends appear to be gradually increasing over time. Previous studies have suggested that only a fraction of patients with SRMs are managed with EM in the United States. For instance, Cooperberg et al. [24] also used the NCDB to elucidate the national treatment trends for clinical T1a renal
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
5
Table 2 Multivariable logistic regression analysis in the use of different treatments and EM for localized renal cell carcinoma by tumor size among patients aged 70 to 79 years stratified by tumor size Covariate (referent)
OR (95% CI)
PN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2001
o2 cm
RN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 Ablation Hospital type (community) Comprehensive community Academic Time Interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 EM Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011
P value
OR (95% CI)
P value
2–3.9 cm
OR (95% CI)
P value
4–7 cm
1.49 (1.08–2.0) 1.73 (1.25–2.40)
0.01 0.001
1.23 (0.90–1.15) 1.91 (1.64–2.29)
0.73 o0.001
1.39 (1.06–1.82) 2.74 (2.10–3.60)
0.02 o0.001
1.07 1.07 0.97 1.44
0.69 0.67 0.85 0.03
0.91 1.01 1.32 1.87
(0.77–1.08) (0.86–1.19) (1.12–1.56) (1.59–2.20)
0.33 0.87 0.001 o0.001
1.09 1.55 2.37 3.17
(0.80–1.49) (1.15–2.08) (1.78–3.16) (2.40–4.21)
0.64 0.003 o0.001 o0.001
(0.76–1.51) (0.77–1.51) (0.69–1.35) (1.03–2.01)
0.72 (0.52–0.99) 0.48 (0.34–0.67)
0.05 o0.001
0.84 (0.74–0.97) 0.49 (0.42–0.51)
0.02 o0.001
0.97 (0.81–1.17) 0.60 (0.50–0.73)
0.82 o0.001
0.65 0.40 0.34 0.24
0.02 o0.001 o0.001 o0.001
0.81 0.62 0.43 0.30
(0.70–0.95) (0.54–0.73) (0.37–0.50) (0.26–0.35)
0.009 o0.001 o0.001 o0.001
0.78 0.58 0.44 0.33
0.04 o0.001 o0.001 o0.001
1.46 (1.15–1.85) 2.18 (1.71–2.79)
0.002 o0.001
0.98 (0.53–1.78) 1.78 (0.98–3.25)
2.46 4.28 5.72 5.42
(1.65–3.49) (2.98–6.14) (4.00–8.17) (3.79–7.74)
o0.001 o0.001 o0.001 o0.001
3.10 4.98 5.44 9.26
0.001 o0.001
0.75 (0.57–0.96) 0.65 (0.50–0.86)
0.02 0.002
1.35 1.52 1.55 1.57
0.08 0.01 0.01 0.008
(0.46–0.93) (0.28–0.57) (0.24–0.48) (0.17–0.35)
1.20 (0.75–1.92) 1.54 (0.95–2.49) 3.79 (1.33–10.80) 8.72 (3.15–24.13) 11.37 (4.14–31.20) 9.91 (3.59 –28.29)
0.77 0.07 0.01 o0.001 o0.001 o0.001
0.65 (0.41–1.03) 0.73 (0.45–1.16)
0.07 0.18
0.70 (0.56–0.87) 0.84 (0.50–0.79)
1.68 2.16 2.59 2.23
0.15 0.03 0.006 0.02
1.33 1.23 1.31 1.61
(0.82–3.45) (1.07–4.36) (1.30–5.13) (1.11–4.45)
tumors. From 1993 to 2007, only 3% to 5% of patients with SRMs were managed with EM. Our study found that more patients were managed nonsurgically, and this appeared to be gradually increasing over time, though it is still made up of a fraction of the entire older patient population. These results are in stark contrast to the growing evidence that EM is a safe option for many patients with SRMs, in particular, patients who are too frail to undergo primary therapy owing to advanced age or comorbidities. In a recent systematic review and meta-analysis of EM, Smaldone et al. [11] reported that approximately a fifth of 880 patients with SRMs remained static (or did not grow) with average growth rate of 0.31 cm per year. Moreover, only 18 patients progressed to distant metastasis (2.3%). In the largest prospective cohort, a multicenter phase 2 trial of 178 patients managed by EM also showed that only 1.1%
(0.99–1.81) (0.91–1.66) (0.97–1.76) (1.21–2.15)
0.06 0.16 0.06 0.001
(0.62–0.98) (0.46–0.73) (0.35–0.54) (0.27–0.42)
(0.91–10.48) (1.52–16.23) (1.67–17.69) (2.89–29.62)
(0.96–1.90) (1.08–2.12) (1.10–2.16) (1.12–2.20)
0.91 0.06 0.06 0.08 0.005 o0.001
(n ¼ 2) patients progressed to metastatic disease as well [12]. In light of these studies and previous studies suggesting marginal benefit of aggressive surgery among older patients with limited life expectancy, our results suggest that there may be some overtreatment of SRMs in this older patient population. Increased attention and incentives to the benefits of EM by avoiding surgery among older patients who are more sensitive to adverse events and the indolent nature of many SRMs may reduce overtreatment with PN and RN. Third, our study also demonstrated that certain hospital characteristics were associated with the use of aggressive surgical treatment for SRMs. Patients surgically treated at academic hospitals were less likely to use EM. To the best of our knowledge, our study is the first to report such a finding where academic hospitals were more likely to treat
6
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
Table 3 Multivariable logistic regression analysis in the use of different treatments and EM with localized renal cell carcinoma by tumor size among patients aged Z80 years stratified by tumor size Covariate (referent)
OR (95% CI)
PN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011
o2 cm
RN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 Ablation Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 EM Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011
P value
OR (95% CI) 2–3.9 cm
2.14 (1.02–4.52) 2.40 (1.11–5.21)
0.04 0.03
1.23 (0.91–1.66) 2.07 (1.52–2.81)
1.07 1.07 0.97 1.44
0.69 0.67 0.85 0.03
1.09 0.86 1.08 1.48
(0.76–1.51) (0.77–1.51) (0.69–1.35) (1.03–2.01)
P value
OR (95% CI)
P value
4–7 cm 0.17 o0.001
1.62 (0.97–2.71) 3.30 (1.98–5.51)
0.10 o0.001 0.80 0.51 0.14 o0.001
(0.77–1.53) (0.61–1.19) (0.77–1.50) (1.07–2.06)
0.60 0.38 0.64 0.02
1.07 1.19 1.48 2.04
(0.61–1.88) (0.69–2.05) (0.87–2.53) (1.20–3.45)
0.42 0.06
1.17 (0.95–1.43) 1.09 (0.87–1.36)
0.54 (0.32–0.94) 0.31 (0.17–0.57)
0.02 o0.001
0.82 (0.89–1.30) 0.82 (0.67–1.01)
0.66 0.35 0.24 0.23
0.21 0.002 o0.001 o0.001
0.73 0.65 0.41 0.38
(0.58–0.93) (0.51–0.82) (0.33–0.52) (0.30–0.48)
0.01 o0.001 o0.001 o0.001
0.76 0.70 0.64 0.53
0.02 o0.001
1.24 (0.92–1.65) 2.22 (1.63–2.97)
0.14 o0.001
1.06 (0.56–1.82) 1.91 (1.05–3.49)
0.96 0.03
0.35 0.008 0.007 0.06
1.84 3.19 4.88 4.48
(1.14–2.98) (2.01–5.06) (3.10–7.68) (2.84–7.06)
0.01 o0.001 o0.001 o0.001
2.39 2.98 4.85 4.97
0.11 0.04 0.003 0.002
0.75 (0.44–1.27) 0.61 (0.35–1.09)
0.29 0.10
0.74 (0.61–0.91) 0.45 (0.36–0.57)
0.005 o0.001
0.73 (0.59–0.91) 0.50 (0.39–0.65)
2.28 2.74 3.59 3.93
0.05 0.01 0.002 0.001
1.14 1.12 1.25 1.20
(0.34–1.26) (0.18–0.67) (0.12–0.47) (0.12–0.46)
3.92 (1.15–13.28) 9.85 (2.89–33.59) 1.88 5.43 5.77 3.47
(0.49–8.12) (1.54–19.06) (1.63–20.53) (0.97–12.37)
(0.99–5.26) (1.21–6.18) (1.58–8.13) (1.74–8.85)
older patients with SRMs. This finding may be attributable to the greater number of resources at high volume academic hospitals where they are more likely to have robotic surgical systems. In a recent population-level study of merged data from the Nationwide Inpatient Sample and the American Hospital Association, the presence of robotic surgery was associated with greater use of PN and more likely to be available at academic medical centers [20]. Using SEER-Medicare from 2001 to 2010, Chandra et al also demonstrated a similar finding that robotic surgical systems increased the use of PN for SRM among Medicare beneficiaries [25]. Identifying other key hospital and surgeon characteristics associated with surgical therapy among older patients with indolent SRMs is needed to limit bias and facilitate more informed decision-making for patients, providers, and key stakeholders.
(0.53–1.50) (0.86–1.47) (0.97–1.63) (0.92–1.57)
0.34 0.37 0.09 0.18
1.24 1.28 1.21 1.33
(0.58–1.00) (0.54–0.92) (0.49–0.84) (0.40–0.69)
(0.82–7.00) (1.04–8.51) (1.73–13.58) (1.77–13.96)
(0.91–1.68) (0.95–1.72) (0.89–1.63) (0.98–1.80)
0.12 0.42 0.05 0.01 0.001 o0.001
0.006 o0.001 0.16 0.10 0.20 0.06
It is important to highlight some of the limitation of our analysis. Our data source does not permit assessment of the decision-making process of patients and providers, renal function, specific comorbidities, and performance status that would have allowed for a more critical analysis of indication of surgery or ablation. Another consideration is that patient preference may have accounted for some of the low adoption of EM, especially considering that it remains a relatively new disease management strategy with limited studies having long-term follow-up greater than 10 years. As the designation of active surveillance (EM) was recently added to the NCDB, our study was limited in determining the national trends of EM. We also acknowledge that some of the patients in the cohort may have been appropriately treated with RN and PN as life expectancy is largely contingent on age, functional status, and comorbidities. Lastly, renal mass biopsy represents another viable disease management strategy for SRM,
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
in particular, for patients who may have limited life expectancy attributable to multiple comorbidities or advanced age [6,26,27]. Unfortunately, the NCDB does not provide a reliable means to identify patients who underwent a renal mass biopsy at the time of diagnosis. 5. Conclusion In summary, most patients who have advanced age and T1 renal tumors with localized RCC are receiving surgery with PN or RN from 2002 to 2011. Over time, there have been shifts toward increased use of PN and EM/active surveillance (AS), though the latter represents a small fraction of all older patients diagnosed with T1 RCC. Increased attention to all available treatment options and EM is needed to facilitate shared decisionmaking about the associated risks and benefits for older patients and their providers such that they make a fully informed treatment decision about SRMs. References [1] Chow WH, Devesa SS, Warren JL, Fraumeni JF Jr. Rising incidence of renal cell cancer in the United States. J Am Med Assoc 1999;281:1628–31. [2] Hollingsworth JM, Miller DC, Daignault S, Hollenbeck BK. Rising incidence of small renal masses: a need to reassess treatment effect. J Natl Cancer Inst 2006;98:1331–4. [3] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 2015;65:5–29. [4] Ortman JM, Velkoff VA, Hogan H. An aging nation: the older population in the United States. Washington, DC: U.S. Census Bureau. Report No.: current population reports; 2014. p. 25-1140. [5] Campbell SC, Novick AC, Belldegrun A, et al. Guideline for management of the clinical T1 renal mass. J Urol 2009;182:1271–9. [6] Motzer RJ, Jonasch E, Agarwal N, et al. Kidney cancer, version 3. J Natl Compr Canc Netw 2015;13:151–9. [7] Kim SP, Thompson RH, Boorjian SA, et al. Comparative effectiveness for survival and renal function of partial and radical nephrectomy for localized renal tumors: a systematic review and meta-analysis. J Urol 2012;188:51–7. [8] Kim SP, Leibovich BC, Shah ND, et al. The relationship of postoperative complications with in-hospital outcomes and costs after renal surgery for kidney cancer. BJU Int 2013;111:580–8. [9] Van Poppel H, Da Pozzo L, Albrecht W, et al. A prospective randomized EORTC intergroup phase 3 study comparing the complications of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol 2007;51:1606–15.
7
[10] Kim SP, Shah ND, Weight CJ, et al. Contemporary trends in nephrectomy for renal cell carcinoma in the United States: results from a population based cohort. J Urol 2011;186:1779–85. [11] Smaldone MC, Kutikov A, Egleston BL, et al. Small renal masses progressing to metastases under active surveillance: a systematic review and pooled analysis. Cancer 2012;118:997–1006. [12] Jewett MA, Mattar K, Basiuk J, et al. Active surveillance of small renal masses: progression patterns of early stage kidney cancer. Eur Urol 2011;60:39–44. [13] Abouassaly R, Lane BR, Novick AC. Active surveillance of renal masses in elderly patients. J Urol 2008;180:505–8:[discussion 8-9]. [14] National Cancer Data Base, Available at: http://ncdbpuf.facs.org/; 2014. [15] Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992;45:613–9. [16] CoC Accreditation Categories—National Cancer Database, Available at: https://www.facs.org/quality-programs/cancer/accredited/about/cat egories; 2015 [accessed 25.06.15]. [17] Bilimoria KY, Stewart AK, Winchester DP, Ko CY. The National Cancer Data Base: a powerful initiative to improve cancer care in the United States. Ann Surg Oncol 2008;15:683–90. [18] StataCorp. Stata Statistical Software. StataCorp LP; 2009. [19] Ma J, Ward EM, Siegel RL, Jemal A. Temporal trends in mortality in the united States, 1969-2013. J Am Med Assoc 2015;314:1731–9. [20] Kardos SV, Gross CP, Shah ND, et al. Association of type of renal surgery and access to robotic technology for kidney cancer: results from a population-based cohort. BJU Int 2014;114:549–54. [21] Tan HJ, Norton EC, Ye Z, Hafez KS, Gore JL, Miller DC. Long-term survival following partial vs radical nephrectomy among older patients with early-stage kidney cancer. J Am Med Assoc 2012;307:1629–35. [22] Kutikov A, Egleston BL, Wong YN, Uzzo RG. Evaluating overall survival and competing risks of death in patients with localized renal cell carcinoma using a comprehensive nomogram. J Clin Oncol 2010;28:311–7. [23] Sun M, Becker A, Tian Z, et al. Management of localized kidney cancer: calculating cancer-specific mortality and competing risks of death for surgery and nonsurgical management. Eur Urol 2014; 65:235–41. [24] Cooperberg MR, Mallin K, Kane CJ, Carroll PR. Treatment trends for stage I renal cell carcinoma. J Urol 2011;186:394–9. [25] Chandra A, Snider JT, Wu Y, Jena A, Goldman DP. Robot-assisted surgery for kidney cancer increased access to a procedure that can reduce mortality and renal failure. Health Aff 2015;34:220–8. [26] Halverson SJ, Kunju LP, Bhalla R, et al. Accuracy of determining small renal mass management with risk stratified biopsies: confirmation by final pathology. J Urol 2013;189:441–6. [27] Kutikov A, Smaldone MC, Uzzo RG, Haifler M, Bratslavsky G, Leibovich BC. Renal mass biopsy: always, sometimes, or never? Eur Urol 2016;70:403–6.
Original article
National treatment trends among older patients with T1-localized renal cell carcinoma1 Simon P. Kim, M.D., M.P.H.a,b,c,*, Cary P. Gross, M.D.c,d, Neal Meropol, M.D.b,e, Alexander Kutikov, M.D.f, Marc C. Smaldone, M.D.f, Nilay D. Shah, Ph.D.g, James B. Yu, M.D.h, Sarah Psutka, M.D., M.S.i, Jonathon Kiechle, M.D.a, Robert Abouassaly, M.D., M.S.a a
Urology Institute, Center of Outcomes and Health Care Quality, University Hospitals Cleveland Medical Center, Cleveland, OH b Seidman Cancer Center, University Hospitals Cleveland Medical Center, Cleveland, OH c Cancer Outcomes Public Policy and Effectiveness Research (COPPER), Yale University, New Haven, CT d Department of Medicine, Yale University, New Haven, CT e Case Comprehensive Cancer Center, Case Western Reserve University, Cleveland, OH f Division of Urologic Oncology, Philadelphia, PA g Division of Health Care Policy and Research, Mayo Clinic, Rochester, MN h Department of Radiation Oncology, Yale University, New Haven, CT i Division of Urologic Surgery, Cook County Health and Hospitals Systems, Chicago, IL Received 3 July 2016; received in revised form 30 September 2016; accepted 7 October 2016
Abstract Objective: To assess the national trends in treatment of localized renal tumors among older patients with limited life expectancy. Materials and methods: Using the National Cancer Database, we identified older patients (Z70 y) diagnosed with T1 renal cell carcinoma from 2002 to 2011. Primary outcome was the initial treatment—partial nephrectomy (PN), radical nephrectomy, EM, and ablation. Multivariable logistic regression analysis stratified by tumor size (o2, 2–3.9, or 4–7 cm) and age groups (70–79 and Z80 y) was used to identify covariates associated with different treatments. Results: Among 41,518 older patients with T1 renal cell carcinoma renal tumors, most were treated with radical nephrectomy (59.0%) followed by PN (20.0%) and ablation (8.4%). Only 12.6% were managed by EM. Among older patients aged 70 to 79 years with renal tumors 2 to 3.9 cm, PN was used more frequently in 2008 to 2009 (odds ratio [OR] ¼ 1.32; P ¼ 0.001) and 2010 to 2011 (OR ¼ 1.87; P o 0.001) compared to 2002 to 2003 and at academic hospitals (OR ¼ 1.91; P o 0.001) compared to community hospitals. Similar trends were observed for patients aged 70 to 79 years with 4 to 7 cm tumors and for patients aged Z80 years across renal tumor sizes. Conclusions: Among older patients with localized renal tumors and limited life expectancy, most are treated surgically with a growing use of PN. A smaller proportion of older patients are managed by EM in the United States. r 2016 Elsevier Inc. All rights reserved.
Keywords: Advanced age; Kidney cancer; Nephrectomy; Renal cell carcinoma; Treatment decisions
1. Introduction Two population trends are occurring in the United States that would increase complexity of treatment decisions for small 1 Dr. Simon P. Kim is supported by a career development award from the Conquer Cancer Foundation from the American Society of Clinical Oncology. * Corresponding author. Fax: þ1-216-844-1100. E-mail address: [email protected] (S.P. Kim).
http://dx.doi.org/10.1016/j.urolonc.2016.10.008 1078-1439/r 2016 Elsevier Inc. All rights reserved.
renal masses (SRMs). The incidence of SRMs has been gradually rising such that clinical T1 renal tumors represent most incident cases, which has been attributable to the growing use of imaging [1,2]. Approximately 60,000 patients would face a diagnosis of renal cell carcinoma (RCC) making it the sixth most common cancer in the United States this year [3]. Against this backdrop, the U.S. population is growing older. The average life expectancies for both men and women have been steadily increasing such that a fifth of the U.S. population would be aged 65 years or older by 2030 [4]. As a result, the rising
2
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
proportion of elderly patients with incidentally detected localized renal tumors with uncertain malignant potential represents a management challenge with significant health policy implications. Clinical practice guidelines currently endorse partial nephrectomy (PN) for SRMs amenable to surgical resection [5–7]. Other possible treatment options for SRMs include radical nephrectomy (RN), ablation, or expectant management (EM). Although PN has become more technically feasible and safer with minimally invasive surgery, such as robotic PN, it still carries some morbidity and bleeding risks that may be poorly tolerated among older patients [8,9]. With the changes in clinical practice guidelines and rapid dissemination of robotic surgery, the national rates of PN have been gradually rising in the United States [10]. However, EM has become an increasingly accepted disease management strategy for SRMs based on the growing number of studies suggesting the modest annual growth rates of the renal tumor and low risk of metastatic potential and cancer-related mortality, in particular among patients with a limited life expectancy [11–13] Yet, an important knowledge gap is the contemporary trends in the treatment of SRMs among older patients in the United States, as some older patients may be exposed to aggressive treatments and its associated morbidities without receiving a survival benefit. In this context, we assessed the contemporary national treatment trends of older patients (470 y) diagnosed with SRMs. 2. Material and methods 2.1. Data source We queried the National Cancer Database (NCDB), a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society, to examine surgical treatment in elderly patients with SRMs [14]. The NCDB is a national oncology dataset obtained from more than 1,500 hospitals that contains no patient- or physician-identifying information. Approximately 70% of newly diagnosed malignancies are captured in the NCDB. 2.2. Study population We identified all adult patients aged Z70 with kidney masses identified as r7 cm or stage T1a or T1b from the years 2002 to 2011. Patients were only included if RCC was their first and solitary cancer diagnosis. Exclusion criteria also included the presence of clinical nodal (N1–3) or metastatic disease (M1), age less than 70 years, and primary tumor in the renal pelvis or ureter. 2.3. Covariates and outcomes Age at diagnosis, race, sex, 2000 census tract annual median income, insurance status (private, Medicare, Medicaid, and
other), geographic region (East, Midwest, West, and South), location (urban, rural, and metro), Charlson-Deyo comorbidity score, education status, and year of treatment were assigned for each patient using NCDB data [15]. Hospital types were stratified using the standard NCDB codes for hospitals. Per NCDB guidelines, academic/research hospitals must be primarily affiliated with a medical school or be a National Cancer Institute-designated comprehensive cancer center. The distinction between comprehensive community and community cancer programs is based on overall case volume and the number of ancillary staff available to help treat oncology patients [16]. The NCDB classifies approximately 20% of institutions as academic/research hospitals, 39% as comprehensive community cancer centers, 35% as community cancer centers, and 6% as other cancer centers [17]. Treatment modalities were defined using NCDB codes for surgical treatment of the primary site to determine receipt of PN, RN, local ablative therapy, or EM. We classified patients who did not receive active treatment as undergoing expectant management. 2.4. Statistical analysis The primary outcome was the initial treatment administered—EM, PN, RN, and ablation. Bivariate associations between treatment type (EM, local ablation, or surgery) and patient characteristics were analyzed using Pearson’s chisquare test. Multivariable logistic regression analysis stratified by tumor size (o2 cm, 2–3.9 cm, or 4–7 cm) and age groups (70–79 and Z80 y) was used to identify covariates associated with different treatments adjusting for patient demographic and clinical characteristics. Stata MP version 11.2 was used to perform all statistical analyses, and a 2-sided P o 0.05 was used to determine statistical significance [18]. 3. Results From 2002 to 2011, we identified 41,518 patients diagnosed with RCC in the NCDB. As shown in Table 1, most patients were aged between 70 and 79 years (71.4%), white (77.3%), and relatively healthy with no Charlson comorbidities (63.0%). Most patients were treated in urban locations (80.8%) and at comprehensive community hospitals (55.3%), whereas only a third of patients were treated at academic hospitals (32.6%). During the study interval, most patients had a clear cell RCC histology (83.0%) with most patients receiving some form of treatment with RN (59.0%), PN (20.0%), and ablation (8.4%). Only 12.6% of patients were managed with EM. Significant temporal trends in the national utilization of different treatments and EM were found from 2002 to 2011 (Fig.). From 2002–2003 to 2010–2011, there was marked decrease in the proportion of patients undergoing RN from 73.5% to 47.3%, respectively, whereas the national utilization of PN rose from 15.2% to 27.4%, respectively (both
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7 Table 1 Clinical and demographic covariates (n ¼ 41,518)
Table 1 Continued
Variable
% (n)
Age (y) 70–79 Z80
71.4 (28,642) 28.6 (11,876)
Female
46.6 (19,363)
Race White Black Hispanic Other
77.3 8.1 11.0 3.6
Charlson comorbidity index 0 1 Z2
63.0 (24,145) 26.2 (10,023) 10.8 (4,132)
Location Urban Metro Rural
80.7 (31,565) 16.8 (6,574) 2.5 (957)
Median zip code income o30,000 30,000–34,999 35,000–45,999 Z46,000
14.1 18.8 29.0 38.1
(5,559) (7,373) (11,399) (14,990)
Health insurance Private Medicaid Medicare Other
10.6 1.1 85.6 2.7
(10.6) (457) (35,552) (1,103)
Geographic region Northeast South Midwest West Hospital type Community Comprehensive community Academic Other
3
(32,101) (3,377) (4,556) (1,484)
21.1 37.4 26.2 15.3 9.7 55.3 32.6 2.4
(4,027) (22,947) (13,530) (1,104)
Location Urban Metro Rural
80.8 (31,565) 16.8 (6,574) 2.4 (957)
Tumor histology Clear cell Papillary Chromophobe Sarcomatoid Other
81.7 10.7 4.7 0.6 2.2
Primary tumor size (cm) o2 2–3.9 Z4
9.6 (3,975) 49.3 (20,472) 41.1 (17,071)
(33,923) (4,469) (1,978) (247) (901)
Variable
% (n)
Primary treatment PN RN Ablation EM
20.0 59.0 8.4 12.6
(8,310) (24,509) (3,458) (5,241)
Time interval 2002–2003 2004–2005 2006–2007 2008–0AS 2010–2011
15.7 18.6 21.5 22.1 22.1
(6,515) (7,708) (8,919) (9,179) (9,197)
P o 0.001 for trend). However, use of renal mass ablation rose from 1.6% in 2002–2003 to 11.7% in 2010–2011. Similarly, the percentage of patients with clinical T1 RCC managed with EM gradually rose during the study interval from 9.8% in 2002–2003 to 13.6% in 2010–2011 (P o 0.001 for trend). On multivariable analysis, our results indicate that the year of treatment and the type of hospital were associated with the primary outcomes when stratified by tumor size and age groups. Among patients aged 70 to 79 years (Table 2), patients had higher odds ratios (OR) for receipt of PN in 2008 to 2009 and 2010 to 2011 for SRMs measuring 2 to 3.9 cm and 4 to 7 cm, respectively, compared with 2002 to 2003 (both P o 0.01). When treated at academic hospitals, patients were also more likely to undergo PN in ORs compared to community hospitals across all different tumor sizes in this age subgroup (all P o 0.05). For instance, patients with T1b renal tumors (4–7 cm) had an OR of 2.74 of undergoing PN at academic hospitals (95% CI: 2.10–3.60). Likewise, RN was associated with decreased use starting in 2004–2005 to 2010–2011 compared to 2002–2003 and when treated at academic hospitals (all P o 0.001). Both ablation and EM also had higher ORs over time and when treated at academic hospitals for larger tumors that were 2.0 to 3.9 and 4 to 7 cm as well (all P o 0.05). Among the subgroup of patients aged Z80 years, the multivariable logistic regression analysis also demonstrated different treatments and EM associated with time intervals and hospital characteristics (Table 3). When compared with 2002 to 2003, PN was also associated with higher OR for renal tumors in 2010 to 2011 measuring o2 cm (OR ¼ 1.44, 95% CI: 1.03–2.01), 2 to 3.9 cm (OR ¼ 1.48, 95% CI: 1.08–2.06), and 4 to 7 cm (OR ¼ 2.04, 95% CI: 1.20–3.45). Likewise, patients who were Z80 years were more likely to undergo PN at academic hospitals across all different tumor size groups compared to community hospitals (all P o 0.05). Older patients were also less likely to undergo RN over more recent years starting from 2006–2007 to 2010–2011 (P o 0.05). Older patients
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
4 80.0% 70.0% 60.0% 50.0%
PN RN Ablation EM
40.0% 30.0% 20.0% 10.0% 0.0%
2002-03
2004-05
2006-07
2008-09
2010-11
Fig. Temporal trends of different treatments and EM over time. All treatments and EM with P o 0.001 for trend.
in this subgroup also had higher OR of undergoing renal mass ablations, however, when treated in more recent years for all time intervals across all tumor sizes (all P o 0.05). Although EM was used more frequently over time with higher ORs from 2006–2007 to 2010–2011 compared to 2002–2003 (all P o 0.05), academic hospitals were associated with statistically significant lower ORs compared to community hospitals for renal tumors sized 2 to 3.9 cm (OR ¼ 0.84, 95% CI: 0.50–0.79) and 4 to 7 cm (OR ¼ 0.65, 95% CI: 0.50–0.86). Likewise, patients who were Z80 years and treated at comprehensive community hospitals were also less likely to undergo EM for renal tumors 2 to 3.9 cm and 4 to 7 cm when compared to community hospitals (both P o 0.05).
4. Discussion Our study presents important information at a time when the patient population is growing older, and there is a rising incidence of SRMs in the United States, along with a growing trend in the use of surgical therapy, in particular, PN [10,19]. Yet, it is essential to recognize that clinical practice guidelines do not identify specific patient characteristics where PN is most efficacious for improving outcomes other than tumor location and complexity. Against this backdrop, our study has several key findings. First, many older patients in the NCDB cohort with clinical T1 RCC received some form of primary therapy (87.4%), where approximately 80% underwent PN or RN. Furthermore, the use of PN nearly doubled with a concomitant drop by a third of RN from 2002 to 2011 among patients aged 470 years. Although previous studies have documented the gradual increases in PN for all patients, we found that older patients are undergoing PN, especially considering the risk of complications with bleeding compared to conventional RN or EM [20]. Indeed, the comparative effectiveness research about whether this segment of the patient population benefits from PN
compared to RN is mixed. Recently, Tan et al. [21] performed an instrumental variable analysis to assess the differences in overall survival and cancer-specific survival among Medicare beneficiaries treated with PN or RN from 1992 to 2007. Although this population-based cohort study demonstrated a lower hazard ratio for all-cause mortality with PN compared to RN (hazard ratio ¼ 0.54, 95% CI: 0.34–0.85) for all Medicare beneficiaries (age 4 65 y), PN did not confer any statistically significant difference in all-cause mortality in subgroup of analysis restricted to patients Z75 years old. In addition, Kutikov et al. [22] developed a nomogram for patients with T1a renal tumors to predict 5- and 10-year noncancer related, kidney cancer-related, and other cancer-related mortality using surveillance, epidemiology, and end results (SEER) data from 1988 to 2003 in order to objectively inform patients about the benefits of treatments. For example, a 75-year-old patient would more likely have a greater risk of non–kidney cancer-related death than RCC mortality at 5-years based on the nomogram (19% vs. 5%). Similarly, Sun et al. [23] also performed a competing risk analysis of Medicare beneficiaries comparing surgical management (PN or RN) vs. nonsurgical management from SEER-Medicare between 1998 and 2005, where patients Z75 years old had negligible survival benefit compared to EM. Taken together, our study suggests that older patients with limited life expectancy are more frequently undergoing PN with a paucity of evidence to support such aggressive treatment and better outcomes. In an era of individualized medicine, shared decision-making and clinical implementation of decision models would facilitate better patient understanding and knowledge about the risks and benefits of all treatment options allowing them to make an informed decision. Second, our study clearly demonstrates that EM is used in most older patients with T1 RCC, though its trends appear to be gradually increasing over time. Previous studies have suggested that only a fraction of patients with SRMs are managed with EM in the United States. For instance, Cooperberg et al. [24] also used the NCDB to elucidate the national treatment trends for clinical T1a renal
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
5
Table 2 Multivariable logistic regression analysis in the use of different treatments and EM for localized renal cell carcinoma by tumor size among patients aged 70 to 79 years stratified by tumor size Covariate (referent)
OR (95% CI)
PN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2001
o2 cm
RN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 Ablation Hospital type (community) Comprehensive community Academic Time Interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 EM Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011
P value
OR (95% CI)
P value
2–3.9 cm
OR (95% CI)
P value
4–7 cm
1.49 (1.08–2.0) 1.73 (1.25–2.40)
0.01 0.001
1.23 (0.90–1.15) 1.91 (1.64–2.29)
0.73 o0.001
1.39 (1.06–1.82) 2.74 (2.10–3.60)
0.02 o0.001
1.07 1.07 0.97 1.44
0.69 0.67 0.85 0.03
0.91 1.01 1.32 1.87
(0.77–1.08) (0.86–1.19) (1.12–1.56) (1.59–2.20)
0.33 0.87 0.001 o0.001
1.09 1.55 2.37 3.17
(0.80–1.49) (1.15–2.08) (1.78–3.16) (2.40–4.21)
0.64 0.003 o0.001 o0.001
(0.76–1.51) (0.77–1.51) (0.69–1.35) (1.03–2.01)
0.72 (0.52–0.99) 0.48 (0.34–0.67)
0.05 o0.001
0.84 (0.74–0.97) 0.49 (0.42–0.51)
0.02 o0.001
0.97 (0.81–1.17) 0.60 (0.50–0.73)
0.82 o0.001
0.65 0.40 0.34 0.24
0.02 o0.001 o0.001 o0.001
0.81 0.62 0.43 0.30
(0.70–0.95) (0.54–0.73) (0.37–0.50) (0.26–0.35)
0.009 o0.001 o0.001 o0.001
0.78 0.58 0.44 0.33
0.04 o0.001 o0.001 o0.001
1.46 (1.15–1.85) 2.18 (1.71–2.79)
0.002 o0.001
0.98 (0.53–1.78) 1.78 (0.98–3.25)
2.46 4.28 5.72 5.42
(1.65–3.49) (2.98–6.14) (4.00–8.17) (3.79–7.74)
o0.001 o0.001 o0.001 o0.001
3.10 4.98 5.44 9.26
0.001 o0.001
0.75 (0.57–0.96) 0.65 (0.50–0.86)
0.02 0.002
1.35 1.52 1.55 1.57
0.08 0.01 0.01 0.008
(0.46–0.93) (0.28–0.57) (0.24–0.48) (0.17–0.35)
1.20 (0.75–1.92) 1.54 (0.95–2.49) 3.79 (1.33–10.80) 8.72 (3.15–24.13) 11.37 (4.14–31.20) 9.91 (3.59 –28.29)
0.77 0.07 0.01 o0.001 o0.001 o0.001
0.65 (0.41–1.03) 0.73 (0.45–1.16)
0.07 0.18
0.70 (0.56–0.87) 0.84 (0.50–0.79)
1.68 2.16 2.59 2.23
0.15 0.03 0.006 0.02
1.33 1.23 1.31 1.61
(0.82–3.45) (1.07–4.36) (1.30–5.13) (1.11–4.45)
tumors. From 1993 to 2007, only 3% to 5% of patients with SRMs were managed with EM. Our study found that more patients were managed nonsurgically, and this appeared to be gradually increasing over time, though it is still made up of a fraction of the entire older patient population. These results are in stark contrast to the growing evidence that EM is a safe option for many patients with SRMs, in particular, patients who are too frail to undergo primary therapy owing to advanced age or comorbidities. In a recent systematic review and meta-analysis of EM, Smaldone et al. [11] reported that approximately a fifth of 880 patients with SRMs remained static (or did not grow) with average growth rate of 0.31 cm per year. Moreover, only 18 patients progressed to distant metastasis (2.3%). In the largest prospective cohort, a multicenter phase 2 trial of 178 patients managed by EM also showed that only 1.1%
(0.99–1.81) (0.91–1.66) (0.97–1.76) (1.21–2.15)
0.06 0.16 0.06 0.001
(0.62–0.98) (0.46–0.73) (0.35–0.54) (0.27–0.42)
(0.91–10.48) (1.52–16.23) (1.67–17.69) (2.89–29.62)
(0.96–1.90) (1.08–2.12) (1.10–2.16) (1.12–2.20)
0.91 0.06 0.06 0.08 0.005 o0.001
(n ¼ 2) patients progressed to metastatic disease as well [12]. In light of these studies and previous studies suggesting marginal benefit of aggressive surgery among older patients with limited life expectancy, our results suggest that there may be some overtreatment of SRMs in this older patient population. Increased attention and incentives to the benefits of EM by avoiding surgery among older patients who are more sensitive to adverse events and the indolent nature of many SRMs may reduce overtreatment with PN and RN. Third, our study also demonstrated that certain hospital characteristics were associated with the use of aggressive surgical treatment for SRMs. Patients surgically treated at academic hospitals were less likely to use EM. To the best of our knowledge, our study is the first to report such a finding where academic hospitals were more likely to treat
6
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
Table 3 Multivariable logistic regression analysis in the use of different treatments and EM with localized renal cell carcinoma by tumor size among patients aged Z80 years stratified by tumor size Covariate (referent)
OR (95% CI)
PN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011
o2 cm
RN Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 Ablation Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011 EM Hospital type (community) Comprehensive community Academic Time interval (2002–2003) 2004–2005 2006–2007 2008–2009 2010–2011
P value
OR (95% CI) 2–3.9 cm
2.14 (1.02–4.52) 2.40 (1.11–5.21)
0.04 0.03
1.23 (0.91–1.66) 2.07 (1.52–2.81)
1.07 1.07 0.97 1.44
0.69 0.67 0.85 0.03
1.09 0.86 1.08 1.48
(0.76–1.51) (0.77–1.51) (0.69–1.35) (1.03–2.01)
P value
OR (95% CI)
P value
4–7 cm 0.17 o0.001
1.62 (0.97–2.71) 3.30 (1.98–5.51)
0.10 o0.001 0.80 0.51 0.14 o0.001
(0.77–1.53) (0.61–1.19) (0.77–1.50) (1.07–2.06)
0.60 0.38 0.64 0.02
1.07 1.19 1.48 2.04
(0.61–1.88) (0.69–2.05) (0.87–2.53) (1.20–3.45)
0.42 0.06
1.17 (0.95–1.43) 1.09 (0.87–1.36)
0.54 (0.32–0.94) 0.31 (0.17–0.57)
0.02 o0.001
0.82 (0.89–1.30) 0.82 (0.67–1.01)
0.66 0.35 0.24 0.23
0.21 0.002 o0.001 o0.001
0.73 0.65 0.41 0.38
(0.58–0.93) (0.51–0.82) (0.33–0.52) (0.30–0.48)
0.01 o0.001 o0.001 o0.001
0.76 0.70 0.64 0.53
0.02 o0.001
1.24 (0.92–1.65) 2.22 (1.63–2.97)
0.14 o0.001
1.06 (0.56–1.82) 1.91 (1.05–3.49)
0.96 0.03
0.35 0.008 0.007 0.06
1.84 3.19 4.88 4.48
(1.14–2.98) (2.01–5.06) (3.10–7.68) (2.84–7.06)
0.01 o0.001 o0.001 o0.001
2.39 2.98 4.85 4.97
0.11 0.04 0.003 0.002
0.75 (0.44–1.27) 0.61 (0.35–1.09)
0.29 0.10
0.74 (0.61–0.91) 0.45 (0.36–0.57)
0.005 o0.001
0.73 (0.59–0.91) 0.50 (0.39–0.65)
2.28 2.74 3.59 3.93
0.05 0.01 0.002 0.001
1.14 1.12 1.25 1.20
(0.34–1.26) (0.18–0.67) (0.12–0.47) (0.12–0.46)
3.92 (1.15–13.28) 9.85 (2.89–33.59) 1.88 5.43 5.77 3.47
(0.49–8.12) (1.54–19.06) (1.63–20.53) (0.97–12.37)
(0.99–5.26) (1.21–6.18) (1.58–8.13) (1.74–8.85)
older patients with SRMs. This finding may be attributable to the greater number of resources at high volume academic hospitals where they are more likely to have robotic surgical systems. In a recent population-level study of merged data from the Nationwide Inpatient Sample and the American Hospital Association, the presence of robotic surgery was associated with greater use of PN and more likely to be available at academic medical centers [20]. Using SEER-Medicare from 2001 to 2010, Chandra et al also demonstrated a similar finding that robotic surgical systems increased the use of PN for SRM among Medicare beneficiaries [25]. Identifying other key hospital and surgeon characteristics associated with surgical therapy among older patients with indolent SRMs is needed to limit bias and facilitate more informed decision-making for patients, providers, and key stakeholders.
(0.53–1.50) (0.86–1.47) (0.97–1.63) (0.92–1.57)
0.34 0.37 0.09 0.18
1.24 1.28 1.21 1.33
(0.58–1.00) (0.54–0.92) (0.49–0.84) (0.40–0.69)
(0.82–7.00) (1.04–8.51) (1.73–13.58) (1.77–13.96)
(0.91–1.68) (0.95–1.72) (0.89–1.63) (0.98–1.80)
0.12 0.42 0.05 0.01 0.001 o0.001
0.006 o0.001 0.16 0.10 0.20 0.06
It is important to highlight some of the limitation of our analysis. Our data source does not permit assessment of the decision-making process of patients and providers, renal function, specific comorbidities, and performance status that would have allowed for a more critical analysis of indication of surgery or ablation. Another consideration is that patient preference may have accounted for some of the low adoption of EM, especially considering that it remains a relatively new disease management strategy with limited studies having long-term follow-up greater than 10 years. As the designation of active surveillance (EM) was recently added to the NCDB, our study was limited in determining the national trends of EM. We also acknowledge that some of the patients in the cohort may have been appropriately treated with RN and PN as life expectancy is largely contingent on age, functional status, and comorbidities. Lastly, renal mass biopsy represents another viable disease management strategy for SRM,
S.P. Kim et al. / Urologic Oncology: Seminars and Original Investigations ] (2016) 1–7
in particular, for patients who may have limited life expectancy attributable to multiple comorbidities or advanced age [6,26,27]. Unfortunately, the NCDB does not provide a reliable means to identify patients who underwent a renal mass biopsy at the time of diagnosis. 5. Conclusion In summary, most patients who have advanced age and T1 renal tumors with localized RCC are receiving surgery with PN or RN from 2002 to 2011. Over time, there have been shifts toward increased use of PN and EM/active surveillance (AS), though the latter represents a small fraction of all older patients diagnosed with T1 RCC. Increased attention to all available treatment options and EM is needed to facilitate shared decisionmaking about the associated risks and benefits for older patients and their providers such that they make a fully informed treatment decision about SRMs. References [1] Chow WH, Devesa SS, Warren JL, Fraumeni JF Jr. Rising incidence of renal cell cancer in the United States. J Am Med Assoc 1999;281:1628–31. [2] Hollingsworth JM, Miller DC, Daignault S, Hollenbeck BK. Rising incidence of small renal masses: a need to reassess treatment effect. J Natl Cancer Inst 2006;98:1331–4. [3] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 2015;65:5–29. [4] Ortman JM, Velkoff VA, Hogan H. An aging nation: the older population in the United States. Washington, DC: U.S. Census Bureau. Report No.: current population reports; 2014. p. 25-1140. [5] Campbell SC, Novick AC, Belldegrun A, et al. Guideline for management of the clinical T1 renal mass. J Urol 2009;182:1271–9. [6] Motzer RJ, Jonasch E, Agarwal N, et al. Kidney cancer, version 3. J Natl Compr Canc Netw 2015;13:151–9. [7] Kim SP, Thompson RH, Boorjian SA, et al. Comparative effectiveness for survival and renal function of partial and radical nephrectomy for localized renal tumors: a systematic review and meta-analysis. J Urol 2012;188:51–7. [8] Kim SP, Leibovich BC, Shah ND, et al. The relationship of postoperative complications with in-hospital outcomes and costs after renal surgery for kidney cancer. BJU Int 2013;111:580–8. [9] Van Poppel H, Da Pozzo L, Albrecht W, et al. A prospective randomized EORTC intergroup phase 3 study comparing the complications of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol 2007;51:1606–15.
7
[10] Kim SP, Shah ND, Weight CJ, et al. Contemporary trends in nephrectomy for renal cell carcinoma in the United States: results from a population based cohort. J Urol 2011;186:1779–85. [11] Smaldone MC, Kutikov A, Egleston BL, et al. Small renal masses progressing to metastases under active surveillance: a systematic review and pooled analysis. Cancer 2012;118:997–1006. [12] Jewett MA, Mattar K, Basiuk J, et al. Active surveillance of small renal masses: progression patterns of early stage kidney cancer. Eur Urol 2011;60:39–44. [13] Abouassaly R, Lane BR, Novick AC. Active surveillance of renal masses in elderly patients. J Urol 2008;180:505–8:[discussion 8-9]. [14] National Cancer Data Base, Available at: http://ncdbpuf.facs.org/; 2014. [15] Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992;45:613–9. [16] CoC Accreditation Categories—National Cancer Database, Available at: https://www.facs.org/quality-programs/cancer/accredited/about/cat egories; 2015 [accessed 25.06.15]. [17] Bilimoria KY, Stewart AK, Winchester DP, Ko CY. The National Cancer Data Base: a powerful initiative to improve cancer care in the United States. Ann Surg Oncol 2008;15:683–90. [18] StataCorp. Stata Statistical Software. StataCorp LP; 2009. [19] Ma J, Ward EM, Siegel RL, Jemal A. Temporal trends in mortality in the united States, 1969-2013. J Am Med Assoc 2015;314:1731–9. [20] Kardos SV, Gross CP, Shah ND, et al. Association of type of renal surgery and access to robotic technology for kidney cancer: results from a population-based cohort. BJU Int 2014;114:549–54. [21] Tan HJ, Norton EC, Ye Z, Hafez KS, Gore JL, Miller DC. Long-term survival following partial vs radical nephrectomy among older patients with early-stage kidney cancer. J Am Med Assoc 2012;307:1629–35. [22] Kutikov A, Egleston BL, Wong YN, Uzzo RG. Evaluating overall survival and competing risks of death in patients with localized renal cell carcinoma using a comprehensive nomogram. J Clin Oncol 2010;28:311–7. [23] Sun M, Becker A, Tian Z, et al. Management of localized kidney cancer: calculating cancer-specific mortality and competing risks of death for surgery and nonsurgical management. Eur Urol 2014; 65:235–41. [24] Cooperberg MR, Mallin K, Kane CJ, Carroll PR. Treatment trends for stage I renal cell carcinoma. J Urol 2011;186:394–9. [25] Chandra A, Snider JT, Wu Y, Jena A, Goldman DP. Robot-assisted surgery for kidney cancer increased access to a procedure that can reduce mortality and renal failure. Health Aff 2015;34:220–8. [26] Halverson SJ, Kunju LP, Bhalla R, et al. Accuracy of determining small renal mass management with risk stratified biopsies: confirmation by final pathology. J Urol 2013;189:441–6. [27] Kutikov A, Smaldone MC, Uzzo RG, Haifler M, Bratslavsky G, Leibovich BC. Renal mass biopsy: always, sometimes, or never? Eur Urol 2016;70:403–6.