Implementation of hypofractionated prostate radiation therapy in the United States: A National Cancer Database analysis

Practical Radiation Oncology (2017) 7, 270–278

CME www.practicalradonc.org

Basic Original Report

Implementation of hypofractionated prostate radiation therapy in the United States: A National Cancer Database analysis William A. Stokes MD ⁎, Brian D. Kavanagh MD, MPH, David Raben MD, Thomas J. Pugh MD ⁎ Department of Radiation Oncology, University of Colorado Denver School of Medicine, Aurora, Colorado Received 24 January 2017; revised 13 March 2017; accepted 31 March 2017

Abstract Purpose: Preclinical and clinical research over the past several decades suggests that hypofractionated (HFxn) radiation therapy schedules produce similar treatment outcomes compared with conventionally fractionated (CFxn) radiation therapy for definitive treatment of localized prostate cancer (PCa). We sought to evaluate national trends and identify factors associated with HFxn utilization using the US National Cancer Database. Methods and materials: We queried the National Cancer Database for men diagnosed with localized (N0,M0) PCa from 2004 through 2013 treated with external beam radiation therapy. Patients were grouped by dose per fraction (DpF) in Gray: CFxn was defined as DpF ≤2.0, moderate HFxn as DpF N2.0 but b5.0, and extreme HFxn as DpF ≥5.0. Men receiving DpF b1.5 or N15.0 were excluded, as were those receiving b25 or N90 Gy total dose. Multiple logistic regression was performed to identify demographic, clinical, and treatment factor associations. Results: A total of 132,403 men were identified, with 120,055 receiving CFxn, 7264 moderate HFxn, and 5084 extreme HFxn. Although CFxn was by far the most common approach over the analysis period, HFxn use increased from 6.2% in 2004 to 14.2% in 2013 (P b .01). Extreme HFxn use increased the most (from 0.3% to 8.5%), whereas moderate HFxn utilization was unchanged (from 5.9% to 5.7%). HFxn use was independently associated with younger age, later year of diagnosis, non-black race, non-Medicaid insurance, non-Western residence, higher income, academic treatment facility, greater distance from treatment facility, low-risk disease group (by National Comprehensive Cancer Network criteria), and nonreceipt of hormone therapy. Conclusions: Although CFxn remains the most common radiation therapy schedule for localized PCa, use of HFxn appears to be increasing in the United States as a result of increased extreme HFxn use. Financial and logistical factors may accelerate adoption of shorter schedules. Considering the multiple demographic and prognostic differences identified between these groups, randomized outcome data comparing extreme HFxn to alternatives are desirable. © 2017 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.

Note—Earn CME credit by taking a brief online assessment at https://www.astro.org/Meeting-and-Education.aspx Conflicts of interest: None. ⁎ Corresponding authors. Department of Radiation Oncology, 1665 Aurora Ct, Suite 1032, MS F706, Aurora, CO 80045. E-mail address: [email protected] (W.A. Stokes), [email protected] (T.J Pugh). http://dx.doi.org/10.1016/j.prro.2017.03.011 1879-8500/© 2017 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.

Practical Radiation Oncology: July-August 2017

Introduction Definitive external-beam radiation therapy (EBRT) constitutes an accepted therapeutic approach in the management of localized prostate cancer (PCa). 1 Although men electing this option have typically received conventional fractionation (CFxn), 2 research over the past few decades has suggested that PCa and surrounding normal tissue respond differently to various fractionation schema, 3-5 thereby stimulating interest in hypofractionated (HFxn) schedules, which entail a shorter course with fewer, larger fractions. A survey of practice patterns among PCa patients who initiated definitive EBRT in 1994 at 80 facilities across the United States identified a modal dose per fraction typical of CFxn 2; however, to our knowledge, no subsequent analysis has been published evaluating fractionation patterns at a national level. Since the publication of these survey results in 2001, emerging data from several institutions have demonstrated the feasibility of moderate HFxn (ModHFxn), 6,7 and, more recently, early reports of randomized trials (including Radiation Therapy Oncology Group [RTOG] 0415, 8 Prostate Fractionated Irridation [PROFIT], 9 and Conventional or Hypofractionated High Dose Intensity Modulated Radiotherapy for Prostate Cancer [CHHiP] 10) have demonstrated biochemical control with ModHFxn that is at least comparable to that offered by CFxn. Technological advances have also spurred interest in extreme HFxn (ExtHFxn), typically delivered in a small number of highly conformal fractions of stereotactic body radiation therapy (SBRT). Despite encouraging early results from series of carefully selected patients undergoing ExtHFxn, 11,12 there are no published randomized data comparing such regimens with ModHFxn or CFxn. Although some have argued that an abbreviated treatment course with logistical and possible financial advantages is an appropriate option for select patients, 13,14 it is unknown how these emerging data on HFxn have affected fractionation schedules for PCa on a national level. We therefore aimed to describe fractionation patterns among American men undergoing definitive EBRT for PCa and to describe factors associated with receipt of HFxn using the National Cancer Database (NCDB).

Materials and methods The NCDB, a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society, is a hospital-based registry capturing approximately 70% of incident cancer cases in the United States and drawing data from more than 1500 Commission-accredited cancer programs. The NCDB contains detailed information on demographic, clinical, and treatment-related factors. The American College of Surgeons and the Commission on Cancer have not verified

Fractionation for localized prostate cancer

271

and are not responsible for the analytic or statistical methodology used or for the conclusions drawn from these data by the investigators. The present analysis was performed with the approval of our local institutional review board. We initially queried the NCDB for cases of prostatic adenocarcinoma diagnosed from 2004 through 2013 in men aged 40 to 90 years who received radiation therapy without surgery, excluding patients with node-positive or metastatic disease and those with a prior diagnosis of malignancy (n = 176,188). We excluded those receiving any radiation modality other than EBRT and those missing complete information for their regional radiation dose, boost radiation dose, and number of fractions (n = 34,699). We then excluded patients coded as receiving palliative interventions (n = 141), those receiving b25 or N90 Gy in total dose (n = 4126), and those receiving b1 or N50 fractions (n = 1470) and divided each subject’s total radiation dose in Gray by the number of fractions to calculate a dose per fraction. To account for potential inaccuracies in recorded dose and fraction number, we broadly defined 3 fractionation cohorts: a CFxn cohort receiving 1.5 to 2.0 Gy per fraction, a ModHFxn cohort receiving N2.0 but b5.0 Gy per fraction, and an ExtHFxn cohort receiving 5.0 to 15.0 Gy per fraction. Men receiving b1.5 or N15.0 Gy per fraction were excluded (n = 3349). Each patient was then classified as having either low-, intermediate-, or high-risk PCa according to his clinical T classification, prostate-specific antigen, and Gleason score, per National Comprehensive Cancer Network criteria. 1 Besides risk group, patient-specific covariates incorporated into our analysis included age, year of diagnosis, race/ ethnicity, insurance status, income (by ZIP code quartile), and comorbidity. We also analyzed facility-specific factors including geographic region (grouped into East, Midwest, South, and West), facility type, and distance from facility. Hormone therapy (HT) was incorporated as a treatmentspecific covariate. Statistical analyses were performed using SPSS V24.0 (SPSS Inc., Chicago, IL). Covariates were selected a priori. Pearson χ 2 tests were used to assess associations between variables and fractionation schedule. Multivariable binary logistic regression models were used to assess the association between fractionation schedule and demographic, clinical, and treatment factors, with the results reported as odds ratios (ORs) for receipt of various fractionation schedules with corresponding 95% confidence intervals (95% CIs). All tests were 2-sided with a .05 level of significance.

Results In total, 132,403 men met inclusion criteria, including 120,055 (90.7%) receiving CFxn, 7264 (5.5%) ModHFxn, and 5084 (3.8%) ExtHFxn. A majority of each fractionation

272

W.A. Stokes et al

group comprised men who were older than 65 years of age and white, had Medicare and comorbidity scores of 0, underwent care at facilities located 30 miles or less from their primary address, and did not receive HT (Table 1). There were significant baseline differences among the fractionation groups, with patients who were older, diagnosed in their earlier years, non-Hispanic, insured by Medicaid, residing in lower income areas in the Midwest, undergoing care at nonacademic facilities located less than 10 miles from their primary address, having higher comorbidity scores and high-risk PCa, and undergoing HT all more likely to receive CFxn (all P b .01). The median CFxn patient received 77.40 Gy (interquartile range [IQR], 75.60-79.20) in 42 fractions (IQR, 40-43); the median ModHFxn patient received 75.00 Gy (IQR, 70.00-77.40) in 28 fractions (IQR, 25-34); and the median ExtHFxn patient received 36.25 Gy (IQR, 36.25-37.50) in 5 fractions (IQR, 5-5). Use of CFxn decreased over time, from 93.8% in 2004 to 85.8% in 2013. This decrease corresponded to growth in the application of ExtHFxn from 0.3% to 8.5% over the same period, rather than any change in use of ModHFxn, which changed only slightly from 5.9% in 2004 to 5.7% in 2013 (Fig 1). Univariate analysis of year as a continuous variable confirmed growing utilization of HFxn (either ModHFxn or ExtHFxn) over time (OR, 1.11; 95% CI, 1.10-1.12; P b .01) (Table 2). Controlling for baseline differences on multivariate analysis, HFxn patients remained more likely to be diagnosed later within the time period studied (OR, 1.10; 95% CI, 1.09-1.11; P b .01). HFxn use was also associated with younger age, non-black race, Medicare or private insurance rather than Medicaid, residence in non-Western regions and higher income areas, receipt of care at academic facilities located farther from their primary residence, low-risk PCa, and nonreceipt of HT (Table 2). Limiting our analytic cohort to men undergoing HFxn, we then performed multivariate regression to identify factors associated with receipt of ExtHFxn versus ModHFxn. Later year of diagnosis was again a significant predictor (OR, 1.35; 95% CI, 1.32-1.38; P b .01), as were age b65 years, white race, Medicare insurance, non-Western location, residence in higher income areas, comorbidity score of 1, receipt of care at a comprehensive community cancer program (OR, 1.33; 95% CI, 1.19-1.49; P b .01) rather than an academic (reference) or community cancer program (OR, 0.40; 95% CI, 0.32-0.51; P b .01), distance greater than 10 miles to treatment facility, low-risk disease rather than intermediate-risk (OR, 0.71; 95% CI, 0.64-0.80, P b .01) or high-risk (OR, 0.36; 95% CI, 0.31-0.42; P b .01) disease, and nonreceipt of HT.

Discussion In the largest study to date of EBRT fractionation for the definitive management of localized PCa in the United States, we identified increased implementation of HFxn

Practical Radiation Oncology: July-August 2017

over the decade through 2013, although CFxn remained by far the most common fractionation schedule. Notably, the growth in HFxn seemed to come entirely from ExtHFxn, with rates of ModHFxn remaining comparatively unchanged over the period studied. Care patterns identified here suggest that providers are generally exercising cautious use of HFxn. In particular, we identified increased use of HFxn among men who live at greater distances from their treatment facility, implying that providers may be offering these patients a shorter regimen with fewer treatments for reasons of logistical convenience. We also identified an inverse correlation between HFxn and risk group, with low-risk patients the most likely to receive HFxn and ExtHFxn. This pattern is consistent with the early published experience with ModHFxn 6,15 and ExtHFxn, 16,17 which has primarily focused on men classified as low or intermediate risk. Moreover, the temporal trends observed in the growth of ExtHFxn seem to reflect the accumulation of published data, with very limited use before the first publication of short-term outcomes from the Stereotactic Hypofractionated Accurate Radiotherapy of the Prostate (SHARP) trial in 2007, 16 followed by growing adoption as more groups published their findings in the ensuing years. 11,17-21 We also identified sociodemographic disparities that persisted even after adjusting for clinical and geographic factors. Notably, black men were less likely than their white counterparts to receive HFxn and, among those receiving HFxn, were less likely to undergo ExtHFxn. This finding adds to the growing body of literature documenting differing care patterns between white and black PCa patients. 22-25 We also found reduced receipt of HFxn, particularly ExtHFxn, among men residing in lower income regions and those with insurance other than Medicare. Altogether, these sociodemographic disparities underscore the importance of reducing barriers to accessing health care. Given the growing interest in HFxn over the past decade, it may come as no surprise that utilization of CFxn has seen a steady decline. It is remarkable, however, that the fractionation schedule taking its place is overwhelmingly ExtHFxn rather than ModHFxn. Multiple schedules of ModHFxn have been compared head-to-head with CFxn in a number of randomized trials, including RTOG 0415, 8 PROFIT, 9 CHHiP, 10 and others, 15,26-30 each yielding at least comparable biochemical control to CFxn. In comparison, the published experience with ExtHFxn consists entirely of single-arm studies 19-21,31-37 and pooled analyses. 11,12 Aside from the recent presentation in abstract form of 1- and 2-year toxicity and quality-of-life outcomes from RTOG 0938 and the Swedish Hypofractionated Radiotherapy of intermediate Risk Localised Prostate Cancer trial (HYPO-RT-PC), 38,39 randomized comparisons with CFxn or ModHFxn are sorely lacking at present. As a result, use of ModHFxn schedules supported by randomized evidence compiled over the

Practical Radiation Oncology: July-August 2017

Table 1

273

Fractionation for localized prostate cancer

Demographics and clinical characteristics Patients (N = 132,403) CFxn (n = 120,055) No.

Age at diagnosis, y 40-54 55-64 65-74 75-90 Year of diagnosis 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Race White Black Asian-Pacific Islander Native American Hispanic Other/unknown Insurance Uninsured Medicaid Medicare Private insurance Other/unknown Geography East Midwest South West Income b$38,000 $38,000-47,999 $48,000-62,999 ≥$63,000 Unknown Charlson-Deyo comorbidity 0 1 2+ Facility type Academic Community cancer Comprehensive community Other/unknown

ModHFxn (n = 7264) %

No.

5028 27,484 56,276 31,267

88.2 89.4 90.8 92.0

344 1818 3307 1795

9916 10,493 12,151 13,722 13,697 13,168 12,924 13,103 10,828 10,053

93.8 93.5 92.6 92.7 91.6 90.9 89.9 88.5 87.6 85.8

90,038 20,216 259 2644 4743 2155

ExtHFxn (n = 5084) %

No.

%

6.0 5.9 5.3 5.3

331 1437 2396 920

5.8 4.7 3.9 2.7

628 665 780 764 779 776 704 839 666 663

5.9 5.9 5.9 5.2 5.2 5.4 4.9 5.7 5.4 5.7

27 61 188 314 477 543 744 859 870 1,001

0.3 0.5 1.4 2.1 3.2 3.7 5.2 5.8 7.0 8.5

90.7 91.5 90.2 90.3 87.2 90.1

5,110 1160 19 210 604 161

5.1 5.2 6.6 7.2 11.1 6.7

4,110 722 9 73 95 75

4.1 3.3 3.1 2.5 1.7 3.1

1959 3484 73,351 36,272 4989

90.4 93.8 91.3 89.4 88.9

141 162 4071 2437 453

6.5 4.4 5.1 6.0 8.1

66 67 2928 1855 168

3.0 1.8 3.6 4.6 3.0

32,086 31,067 37,031 19,331

90.6 91.9 89.9 90.4

1230 1853 2474 1707

3.5 5.4 6.0 8.0

2086 950 1698 350

5.9 2.8 4.1 1.6

21,936 27,789 31,582 37,696 1052

92.1 92.3 91.6 87.9 91.3

1323 1547 1788 2548 58

5.6 5.1 5.2 5.9 5.0

554 775 1090 2623 42

2.3 2.6 3.2 6.1 3.6

104,512 12,811 2732

90.6 90.8 92.2

6403 709 152

5.6 5.0 5.1

4410 595 79

3.8 4.2 2.7

36,073 14,619 61,401 7962

85.7 95.0 92.9 89.9

3258 619 2872 515

7.7 4.0 4.3 5.8

2761 146 1802 375

6.6 0.9 2.7 4.2

P b.01

b.01

b.01

b.01

b.01

b.01

b.01

b.01

(continued on next page)

274

W.A. Stokes et al

Practical Radiation Oncology: July-August 2017

Table 1 (continued) Patients (N = 132,403) CFxn (n = 120,055) No. Distance, miles b10 10-30 N30 Unknown NCCN risk group Low Intermediate High Unknown Hormone therapy No Yes Unknown

ModHFxn (n = 7264) %

No.

68,411 35,401 15,330 913

91.9 90.4 86.2 91.9

3936 2058 1221 49

29,598 52,594 36,299 1564

88.9 90.6 92.3 91.8

61,002 55,999 3054

88.4 93.3 89.9

ExtHFxn (n = 5084) %

No.

%

5.3 5.3 6.9 4.9

2119 1701 1232 32

2.8 4.3 6.9 3.2

1708 3052 2409 95

5.1 5.3 6.1 5.6

2000 2410 629 45

6.0 4.2 1.6 2.6

3745 3395 124

5.4 5.7 3.7

4223 642 219

6.1 1.1 6.4

P b.01

b.01

b.01

CFxn, conventional fractionation; ExtHFxn, extreme hypofractionation; ModHFxn, moderate hypofractionation, NCCN, National Comprehensive Cancer Network.

past decade has not changed, whereas implementation of the less rigorously studied ExtHFxn schedules has increased significantly. Because the application of evidence-based medicine does not appear to account for these adoption patterns, it is reasonable to consider nonclinical factors. Among these, varying clinical experience with HFxn across different facility types deserves particular consideration. Our first logistic regression model (Table 2) assessing factors associated with receipt of HFxn (primarily ModHFxn) versus CFxn identified academic facility type as having the

Figure 1 Fractionation schedules of definitive radiation therapy for prostate cancer, 2004-2013. CFxn, conventional fractionation; ExtHFxn, extreme hypofractionation; ModHFxn, moderate hypofractionation.

strongest association with HFxn. However, our second logistic regression model (Table 3), which assessed predictors of receiving ExtHFxn versus ModHFxn, demonstrated increased use of ExtHFxn at comprehensive cancer programs and reduced utilization at community cancer programs as compared with academic facilities. One explanation for these findings is that formal National Cancer Institute (NCI)–sponsored studies of ModHFxn were initiated several years before the first NCI–sponsored studies of ExtHFxn, and so expertise with and supporting evidence for ModHFxn at academic centers predated a similar comfort level with ExtHFxn. Interestingly, although small formal prospective studies of ExtHFxn were launched at 2 academic centers more than 15 years ago, 40,41 the majority of experience with ExtHFxn accumulated in larger community-based centers in more of a registry-type repository rather than hypothesis-driven clinical trials. Ultimately, our findings imply that both changes to reimbursement for various radiation therapeutic approaches and ongoing dissemination of technological expertise in SBRT may have significant implications for future care patterns among American PCa patients. In the meantime, long-term efficacy and toxicity outcomes of randomized trials comparing ExtHFxn with CFxn or ModHFxn are eagerly awaited. Although the current analysis is the first to assess patterns of EBRT in PCa with respect to fractionation schedules, a recent analysis of the NCDB by Baker and colleagues 42 examined dissemination of SBRT for the definitive treatment of localized PCa. Although both analyses depict adoption of HFxn EBRT across the United States starting in 2004, some important distinctions are

Practical Radiation Oncology: July-August 2017 Table 2

275

Fractionation for localized prostate cancer

Predictors of HFxn (vs CFxn)

Variable

Age at diagnosis, y (vs 40-54) 55-64 65-74 75-90 Year of diagnosis Continuous Race (vs white) Black Asian-Pacific Islander Native American Hispanic Insurance (vs uninsured) Medicaid Medicare Private insurance Geography (vs West) Midwest South East Income (vs b$38,000) $38,000-47,999 $48,000-62,999 ≥$63,000 Charlson-Deyo Comorbidity (vs 0) 1 2+ Facility type (vs academic) Community cancer Comprehensive community Other Distance (vs b10 miles) 10-30 miles N30 miles NCCN risk group (vs low) Intermediate High Hormone therapy (vs no) Yes

Univariate analysis

Multivariate analysis

OR

95% CI

P

OR

95% CI

P

0.88 0.76 0.65

0.81-0.96 0.69-0.82 0.59-0.71

.01 b.01 b.01

0.87 0.81 0.79

0.79-0.96 0.73-0.89 0.71-0.88

b.01 b.01 b.01

1.11

1.10-1.12

b.01

1.10

1.09-1.11

b.01

0.91 1.06 1.05 1.44

0.86-0.96 0.71-1.56 0.92-1.18 1.33-1.56

b.01 .79 .49 b.01

0.90 1.02 1.15 1.72

0.84-0.95 0.65-1.61 1.01-1.32 1.57-1.88

b.01 .93 .04 b.01

0.62 0.90 1.12

0.51-0.76 0.78-1.04 0.97-1.30

b.01 .17 .13

0.77 1.21 1.24

0.63-0.95 1.03-1.42 1.06-1.45

.02 .02 .01

0.83 1.06 0.97

0.79-0.89 1.00-1.12 0.92-1.03

b.01 .04 .32

1.15 1.59 1.14

1.07-1.23 1.49-1.69 1.07-1.22

b.01 b.01 b.01

0.98 1.07 1.60

0.92-1.04 1.00-1.13 1.52-1.69

.46 .04 b.01

1.02 1.10 1.57

0.95-1.09 1.03-1.18 1.47-1.68

.67 .01 b.01

0.98 0.82

0.93-1.05 0.71-0.94

.60 b.01

1.03 0.89

0.97-1.10 0.77-1.03

.33 .13

0.31 0.46 0.67

0.29-0.34 0.44-0.48 0.62-0.72

b.01 b.01 b.01

0.38 0.51 0.73

0.35-0.42 0.49-0.53 0.67-0.79

b.01 b.01 b.01

1.20 1.81

1.15-1.25 1.72-1.90

b.01 b.01

1.11 1.60

1.06-1.17 1.51-1.70

b.01 b.01

0.83 0.67

0.79-0.87 0.64-0.70

b.01 b.01

0.91 0.90

0.86-0.95 0.84-0.96

b.01 b.01

0.55

0.53-0.57

b.01

0.62

0.59-0.65

b.01

CFxn, conventional fractionation; CI, confidence interval; HFxn, hypofractionation; NCCN, National Comprehensive Cancer Network; OR, odds ratio.

worth noting. First, the earlier study is focused on dissemination of ExtHFxn/SBRT alone, whereas the present analysis also examines adoption of ModHFxn. Moreover, the 2 studies segment their cohorts differently, with the earlier study focusing on radiation technology as coded in 1 part of NCDB data and the present analysis examining fractionation schedules according to dose-per-fraction calculated from a different part of the available data. Because of potentially incomplete or miscoded data, these distinct classification schema may result in slightly different ExtHFxn/SBRT groups. Despite these differences, the analyses identify similar patterns in the adoption of ExtHFxn. Both find increased use of

ExtHFxn/SBRT since 2004, but particularly among men who are white, have Medicare, reside in higher income areas in non-Western parts of the country, undergo care at academic or comprehensive community facilities, and have an intermediate comorbidity burden and low-risk PCa. Our study joins other recent large registry-based analyses of nationwide patterns of radiation therapy fractionation. Consistent with earlier studies of adjuvant radiation therapy among women with early-stage breast cancer 43,44 and of palliative radiation therapy among patients with bone metastases, 45 we identified greater use of shorter courses at academic facilities and with the

276

W.A. Stokes et al

Table 3

Practical Radiation Oncology: July-August 2017

Predictors of ExtHFxn (vs ModHFxn)

Variable

Age at diagnosis, y (vs 40-54) 55-64 65-74 75-90 Year of diagnosis Continuous Race (vs white) Black Asian-Pacific Islander Native American Hispanic Insurance (vs uninsured) Medicaid Medicare Private insurance Geography (vs West) Midwest South East Income (vs b$38,000) $38,000-47,999 $48,000-62,999 ≥$63,000 Charlson-Deyo comorbidity (vs 0) 1 2+ Facility type (vs academic) Community cancer Comprehensive community Other Distance (vs b10 miles) 10-30 miles N30 miles NCCN risk group (vs low) Intermediate High Hormone therapy (vs no) Yes

Univariate analysis

Multivariate analysis

OR

95% CI

P

0.82 0.75 0.53

0.70-0.97 0.64-0.88 0.45-0.63

.02 b.01 b.01

1.33

1.31-1.35

0.77 0.59 0.43 0.20

OR

95% CI

P

0.90 0.78 0.71

0.72-1.12 0.62-0.99 0.55-0.91

.33 .04 b.01

b.01

1.35

1.32-1.38

b.01

0.70-0.86 0.27-1.30 0.33-0.57 0.16-0.24

b.01 .19 b.01 b.01

0.84 0.75 0.85 0.27

0.72-0.96 0.24-2.28 0.59-1.21 0.21-0.35

.01 .61 .63 b.01

0.88 1.54 1.63

0.59-1.33 1.14-2.07 1.21-2.19

.55 b.01 b.01

0.93 1.56 1.09

0.55-1.56 1.06-2.29 0.75-1.59

.77 .02 .66

2.50 3.35 8.27

2.18-2.87 2.94-3.81 7.23-9.47

b.01 b.01 b.01

3.28 5.57 11.2

2.76-3.91 4.71-6.58 9.43-13.3

b.01 b.01 b.01

1.20 1.46 2.46

1.05-1.36 1.29-1.65 2.20-2.75

.01 b.01 b.01

1.26 1.38 2.14

1.06-1.50 1.17-1.63 1.83-2.51

.01 b.01 b.01

1.22 0.76

1.09-1.37 0.57-0.99

b.01 .04

1.20 0.73

1.03-1.40 0.52-1.04

.02 .08

0.28 0.74 0.86

0.23-0.34 0.69-0.80 0.75-0.99

b.01 b.01 .04

0.40 1.33 1.19

0.32-0.51 1.19-1.49 0.99-1.44

b.01 b.01 .07

1.54 1.87

1.41-1.67 1.70-2.06

b.01 b.01

1.25 2.19

1.12-1.40 1.92-2.50

b.01 b.01

0.67 0.22

0.62-0.73 0.20-0.25

b.01 b.01

0.71 0.36

0.64-0.80 0.31-0.42

b.01 b.01

0.17

0.15-0.18

b.01

0.22

0.20-0.25

b.01

CI, confidence interval; ExtHFxn, extreme hypofractionation; ModHFxn, moderate hypofractionation; NCCN, National Comprehensive Cancer Network; OR, odds ratio.

passage of time. These commonalities suggest that the observed variations in fractionation may be related to varying degrees of familiarity with the medical literature, temporal dissemination of knowledge, and different reimbursement schema. Turning to the future, we expect use of HFxn to continue to grow from its relatively low rate of approximately 15% in 2013. Based on recent trends, we would anticipate this growth to continue to come from ExtHFxn, particularly with further dissemination of expertise in ExtHFxn across the nation. Our analysis has identified several groups among whom HFxn uptake has been comparatively modest and whose providers may

warrant focused outreach. These include older men, blacks, those residing in low-income areas, and especially those living in the West. Utilization of ModHFxn will likely remain low in the near future, at least until the publication of long-term results from randomized comparisons of HFxn and CFxn. 8-10 In the meantime, 2 factors that may accelerate adoption of HFxn include changes in clinical guidelines 1 from listing HFxn as an option to endorsing it in appropriately selected patients and shifting reimbursement patterns away from fee-for-service and toward bundled payments. Future analyses of national fractionation patterns for PCa should take into account each of these factors.

Practical Radiation Oncology: July-August 2017

As a database analysis, our study has inherent limitations, including the potential for miscoding of variables and selection bias that is not accounted for by variables available in the database. Additionally, factors of potential medical decision-making influence, such as baseline genitourinary symptoms, are not available. Finally, although NCDB captures approximately 70% of incident cancers in the United States, the findings of this analysis may not be generalizable to cases of PCa diagnosed at centers that do not report to NCDB.

Conclusions In the largest analysis to date of fractionation schedules for localized PCa in the United States, utilization of HFxn increased from 2004 through 2013, driven primarily by adoption of ExtHFxn rather than ModHFxn, despite the absence of randomized data supporting oncologic equivalency of ExtHFxn schedules. Although academic facilities have contributed to the growth of HFxn, implementation of ExtHFxn appears to be driven by comprehensive community cancer programs treating men with low-risk disease.

References 1. Mohler JL, Armstrong AJ, Bahnson RR, et al. Prostate cancer, version 1.2016. J Natl Compr Cancer Netw. 2016;14:19-30. 2. Zietman A, Moughan J, Owen J, Hanks G. The patterns of care survey of radiation therapy in localized prostate cancer: Similarities between the practice nationally and in minority-rich areas. Int J Radiat Oncol Biol Phys. 2001;50:75-80. 3. Deore SM, Shrivastava SK, Supe SJ, Viswanathan PS, Dinshaw KA. Alpha/beta value and importance of dose per fraction for the late rectal and recto-sigmoid complications. Strahlenther Onkol. 1993;169:521-526. 4. Haustermans KM, Hofland I, Van Poppel H, et al. Cell kinetic measurements in prostate cancer. Int J Radiat Oncol Biol Phys. 1997;37:1067-1070. 5. Brenner DJ, Martinez AA, Edmundson GK, Mitchell C, Thames HD, Armour EP. Direct evidence that prostate tumors show high sensitivity to fractionation (low α/β ratio), similar to late-responding normal tissue. Int J Radiat Oncol Biol Phys. 2002;52:6-13. 6. Kupelian PA, Willoughby TR, Reddy CA, Klein EA, Mahadevan A. Hypofractionated intensity-modulated radiotherapy (70 gy at 2.5 gy per fraction) for localized prostate cancer: Cleveland clinic experience. Int J Radiat Oncol Biol Phys. 2007;68: 1424-1430. 7. McCammon R, Rusthoven KE, Kavanagh B, Newell S, Newman F, Raben D. Toxicity assessment of pelvic intensity-modulated radiotherapy with hypofractionated simultaneous integrated boost to prostate for intermediate- and high-risk prostate cancer. Int J Radiat Oncol Biol Phys. 2009;75:413-420. 8. Lee WR, Dignam JJ, Amin MB, et al. Randomized phase III noninferiority study comparing two radiotherapy fractionation schedules in patients with low-risk prostate cancer. J Clin Oncol. 2016. JCO670448.

Fractionation for localized prostate cancer

277

9. Catton CN, Lukka H, Gu CS, et al. Randomized trial of a hypofractionated radiation regimen for the treatment of localized prostate cancer. [e-pub ahead of print] J Clin Oncol, http://dx.doi.org/10.1200/JCO.2016.71.7397, accessed April 12, 2017. 10. Dearnaley D, Syndikus I, Mossop H, et al. Conventional versus hypofractionated high-dose intensity-modulated radiotherapy for prostate cancer: 5-year outcomes of the randomised, non-inferiority, phase 3 CHHIP trial. Lancet Oncol. 2016;17:1047-1060. 11. King CR, Freeman D, Kaplan I, et al. Stereotactic body radiotherapy for localized prostate cancer: Pooled analysis from a multi-institutional consortium of prospective phase II trials. Radiother Oncol. 2013;109:217-221. 12. King CR, Collins S, Fuller D, et al. Health-related quality of life after stereotactic body radiation therapy for localized prostate cancer: Results from a multi-institutional consortium of prospective trials. Int J Radiat Oncol Biol Phys. 2013;87:939-945. 13. De Bari B, Arcangeli S, Ciardo D, et al. Extreme hypofractionation for early prostate cancer: Biology meets technology. Cancer Treat Rev. 2016;50:48-60. 14. Lee WR. Hypofractionation for prostate cancer: Tested and proven. Lancet Oncol. 2016;17:1020-1022. 15. Lukka H, Hayter C, Julian JA, et al. Randomized trial comparing two fractionation schedules for patients with localized prostate cancer. J Clin Oncol. 2005;23:6132-6138. 16. Madsen BL, Hsi RA, Pham HT, Fowler JF, Esagui L, Corman J. Stereotactic Hypofractionated Accurate Radiotherapy of the Prostate (SHARP), 33.5 Gy in five fractions for localized disease: First clinical trial results. Int J Radiat Oncol Biol Phys. 2007;67: 1099-1105. 17. Tang C, Loblaw D, Cheung P, et al. Phase I/II study of a five-fraction hypofractionated accelerated radiotherapy treatment for low-risk localised prostate cancer: Early results of PHART3. Clin Oncol. 2008;20:729-737. 18. King CR, Brooks JD, Gill H, Pawlicki T, Cotrutz C, Presti Jr JC. Stereotactic body radiotherapy for localized prostate cancer: Interim results of a prospective phase II clinical trial. Int J Radiat Oncol Biol Phys. 2009;73:1043-1048. 19. Fuller DB, Mardirossian G, Wong D, McKellar H. Prospective evaluation of stereotactic radiotherapy for low and intermediate risk prostate cancer: Emulating HDR brachytherapy dose distribution. Int J Radiat Oncol Biol Phys. 2010;78:S358-S359. 20. Boike TP, Lotan Y, Cho LC, et al. Phase I dose-escalation study of stereotactic body radiation therapy for low- and intermediate-risk prostate cancer. J Clin Oncol. 2011;29:2020-2026. 21. Meier R, Kaplan I, Beckman A, et al. Stereotactic body radiation therapy for intermediate-risk organ-confined prostate cancer: Interim toxicity and quality of life outcomes from a multi-institutional study. Int J Radiat Oncol Biol Phys. 2012;84:S148. 22. Zeliadt SB, Potosky AL, Etzioni R, Ramsey SD, Penson DF. Racial disparity in primary and adjuvant treatment for nonmetastatic prostate cancer: SEER-Medicare trends 1991 to 1999. Urology. 2004;64:1171-1176. 23. Gross CP, Smith BD, Wolf E, Andersen M. Racial disparities in cancer therapy: Did the gap narrow between 1992 and 2002? Cancer. 2008;112:900-908. 24. Moses KA, Paciorek AT, Penson DF, Carroll PR, Master VA. Impact of ethnicity on primary treatment choice and mortality in men with prostate cancer: Data from CaPSURE. J Clin Oncol. 2010;28: 1069-1074. 25. Stokes WA, Hendrix LH, Royce TJ, et al. Racial differences in time from prostate cancer diagnosis to treatment initiation: A population-based study. Cancer. 2013;119:2486-2493. 26. Yeoh EE, Botten RJ, Butters J, Di Matteo AC, Holloway RH, Fowler J. Hypofractionated versus conventionally fractionated radiotherapy for prostate carcinoma: Final results of phase III randomized trial. Int J Radiat Oncol Biol Phys. 2011;81:1271-1278.

278

W.A. Stokes et al

27. Pollack A, Walker G, Horwitz EM, et al. Randomized trial of hypofractionated external-beam radiotherapy for prostate cancer. J Clin Oncol. 2013;1:3860-3868. 28. Hoffman KE, Voong KR, Pugh TJ, et al. Risk of late toxicity in men receiving dose-escalated hypofractionated intensity modulated prostate radiation therapy: Results from a randomized trial. Int J Radiat Oncol Biol Phys. 2014;88:1074-1084. 29. Arcangeli S, Strigari L, Gomellini S, et al. Updated results and patterns of failure in a randomized hypofractionation trial for high-risk prostate cancer. Int J Radiat Oncol Biol Phys. 2012;84: 1172-1178. 30. Incrocci L, Wortel RC, Alemayehu WG, et al. Hypofractionated versus conventionally fractionated radiotherapy for patients with localised prostate cancer (HYPRO): Final efficacy results from a randomised, multicentre, open-label, phase 3 trial. Lancet Oncol. 2016;17:1061-1069. 31. King CR, Brooks JD, Gill H, Presti JC. Long-term outcomes from a prospective trial of stereotactic body radiotherapy for low-risk prostate cancer. Int J Radiat Oncol Biol Phys. 2012;82:877-882. 32. Aluwini S, Beltramo G, Van Rooij P, Boormans J, Kirkels W, Kolkman-Deurloo I-K. P068 stereotactic body radiotherapy with four fractions for low-and intermediate-risk prostate cancer: Acute and late toxicity. Eur Urol Suppl. 2013;6:156. 33. Chen LN, Suy S, Uhm S, et al. Stereotactic body radiation therapy (SBRT) for clinically localized prostate cancer: The Georgetown University experience. Radiat Oncol. 2013;8:1. 34. D'Alimonte L, Loblaw A, Cheung P, et al. Long term outcomes of a novel five fraction hypofractionated protocol for low risk prostate cancer. J Med Imaging Radiat Sci. 2013;44:47. 35. Bolzicco G, Favretto MS, Satariano N, Scremin E, Tambone C, Tasca A. A single-center study of 100 consecutive patients with localized prostate cancer treated with stereotactic body radiotherapy. BMC Urol. 2013;13:49. 36. Katz A, Kang J. Stereotactic body radiation therapy for low-, intermediate-, and high-risk prostate cancer: Disease control and

Practical Radiation Oncology: July-August 2017

37.

38.

39.

40.

41.

42.

43.

44.

45.

quality of life at 6 years. Int J Radiat Oncol Biol Phys. 2013;87: S24-S25. Loblaw A, Cheung P, D’Alimonte L, et al. Prostate stereotactic ablative body radiotherapy using a standard linear accelerator: Toxicity, biochemical, and pathological outcomes. Radiother Oncol. 2013;107:153-158. Lukka H, Stephanie P, Bruner D, et al. Patient-reported outcomes in NRG Oncology/RTOG 0938, a randomized phase 2 study evaluating 2 ultrahypofractionated regimens (UHRs) for prostate cancer. Int J Radiat Oncol Biol Phys. 2016;94:2. Widmark A, Gunnlaugsson A, Beckman L, et al. Extreme hypofractionation versus conventionally fractionated radiotherapy for intermediate risk prostate cancer: Early toxicity results from the Scandinavian randomized phase III trial “HYPO-RT-PC”. Int J Radiat Oncol Biol Phys. 2016;96:938-939. Madsen BL, Hsi RA, Pham HT, et al. Intrafractional stability of the prostate using a stereotactic radiotherapy technique. Int J Radiat Oncol Biol Phys. 2003;57:1285-1291. King CR, Lehmann J, Adler JR, Hai J. CyberKnife radiotherapy for localized prostate cancer: Rationale and technical feasibility. Technol Cancer Res Treat. 2003;2:25-30. Baker BR, Basak R, Mohiuddin JJ, Chen RC. Use of stereotactic body radiotherapy for prostate cancer in the United States from 2004 through 2012. Cancer. 2016;122:2234-2241. Wang EH, Mougalian SS, Soulos PR, et al. Adoption of hypofractionated whole-breast irradiation for early-stage breast cancer: A National Cancer Data Base analysis. Int J Radiat Oncol Biol Phys. 2014;90:993-1000. Jagsi R, Falchook AD, Hendrix LH, Curry H, Chen RC. Adoption of hypofractionated radiation therapy for breast cancer after publication of randomized trials. Int J Radiat Oncol Biol Phys. 2014;90:1001-1009. Rutter CE, James BY, Wilson LD, Park HS. Assessment of national practice for palliative radiation therapy for bone metastases suggests marked underutilization of single-fraction regimens in the United States. Int J Radiat Oncol Biol Phys. 2015;91:548-555.