Prostate Cancer Outcomes by Race Treatment Site

Prostate Cancer Outcomes by Race & Treatment Site Can-lan Sun MD Ph. D, Smita Bhatia MD MPH, Lennie Wong Ph. D, Gail Washington DNS, Karen Nielsen-Menicucci Ph. D 12/11/2008

2008 Estimated US Cancer Deaths* Lung & bronchus 31% Prostate 10% Colon & rectum 8% Pancreas 6% Liver & intrahepatic bile duct 4% Leukemia 4% Esophagus 4% Urinary bladder 3% Non-Hodgkin lymphoma 3% Kidney & renal pelvis 3% All other sites 24% ONS=Other nervous system. Source: American Cancer Society, 2008. Men 294, 120

Cancer Death Rates* by Sex, US, 1975 -2004 Rate Per 100, 000 Men Both Sexes Women *Age-adjusted to the 2000 US standard population. Source: Surveillance, Epidemiology, and End Results (SEER) Program (www. seer. cancer. gov) SEER*Stat Database: Mortality - All COD, Public-Use With State, Total U. S. (1969 -2004), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2007. Underlying mortality data provided by NCHS (www. cdc. gov/nchs).

Cancer Death Rates* Among Men, US, 1930 -2004 Rate Per 100, 000 Prostate *Age-adjusted to the 2000 US standard population. Source: US Mortality Data 1960 -2004, US Mortality Volumes 1930 -1959, National Center for Health Statistics, Centers for Disease Control and Prevention, 2006.

Prostate Cancer Mortality Rates in the US, 1969 -2004

African Americans are twice as likely than Whites to die of prostate cancer. Prostate Cancer Death Rates, 2005 Per 100, 000 population White, Non-Hispanic Deaths Hispanic African American, Non-Hispanic Asian and Pacific Islander American Indian/Alaska Native 6 Note: Data are age adjusted to the 2000 standard population. SOURCE: National Cancer Institute, Surveillence, Epidemiology, and End Results (SEER) Program; National Vital Statistics System--Mortality, NCHS, CDC.

Why? ¢ More aggressive tumors ¢ More advanced stage at diagnosis ¢ Health insurance and access to care ¢ Difference in screening-early detection ¢ Differences in receiving optimal treatment ¢ Socioeconomic status ¢ Healthcare provider

Aims ¢ Aim 1: Compare mortality rates between African-Americans and Caucasians with newly diagnosed prostate cancer in Los Angeles County after controlling for age, SES, marital status, stage, grade, insurance, and treatment modality. ¢ Aim 2: Compare the mortality rates for prostate cancer between NCI designated comprehensive cancer centers and other treatment facilities in Los Angeles County. ¢ Aim 3: Compare the mortality rates by race for patients with prostate cancer receiving care within NCI cancer centers ¢ Aim 4: Describe the proportion of African-Americans and Caucasians seeking treatment for newly diagnosed prostate cancer at NCI designated cancer centers and other treatment facilities, and understand the role of socioeconomic and insurance status in accessing care at the NCI-designated cancer centers versus other treatment facilities

Data Sources ¢ Los Angeles Cancer Surveillance Program (CSP) l l ¢ White or African-American Diagnosed with prostate cancer 1998 -2003 NCI-designated Cancer Center l ¢ USC-Norris Cancer Center, UCLA-Jonsson Cancer Center, City of Hope Cancer Center Office of State Health Planning Department (OSHPD): l Teaching status l Bed Size l Average length of stay l Urban vs. Rural l Hospital in-patient racial distribution l MSSA: % below poverty, racial distribution

Variables ¢ Outcomes l l Prostate cancer-specific mortality l ¢ Overall mortality Time to event (in years from the date of diagnosis to date of death or last known date) Main Exposure l Race: White vs. African-American l NCI designated Cancer Center vs. other non-NCI designated treatment facilities

Variables ¢ Adjustment variables: l Demographics • SES, age, marital status l Year of diagnosis: 1998 -2003 l Insurance l Tumor information • Stage (localized, regional, distant) • Grade (well-differentiated, moderately differentiated, poor/undifferentiated) l Treatment information • Surgery (no, radical/total prostatectomy) • Radiation (yes, no) • Hormone therapy (yes, no)

Preliminary Results 24, 360 Unknown Grade -1474 Unknown Stage -1460 Unknown surgery -1609 22886 21426 Unknown hormone -508 19817 Unknown radiation -11 19309 Unknown Insurance -314 Unknown SES -194 19298 18984 18, 790

Comparison: White vs. AA White N=14, 579 AA N=4211 Alive 11856 (81%) 3328 (79%) Dead 2723 (19%) 883 (21%)* 543 (4%) 200 (5%)* 2180 (15%) 683 (16%)* prostate-specific other *P <0. 05

Comparison: White vs. AA White AA 68 (16 -99) 64 (33 -93)* 12831 (88. 0%) 3708 (88. 1%) 1290 (8. 9%)) 286 (6. 8%)* 458 (3. 1%) 217 (5. 1%)* 525 (3. 6%) 151 (3. 6%) Moderately differentiated 10932 (75. 0%) 3177 (75. 4%) Poor/undifferentiated 3122 (21. 4%) 883 (21. 0%) Age at Dx Stage Local Regional Distant Grade Well-differentiated *P<0. 05

Comparison: White vs. AA White AA Watchful waiting 2284 (15. 7%) 925 (20. 0%)* Surgery only 5419 (37. 2%) 1420 (33. 7%)* Hormone only 1759 (12. 1%) 561 (13. 3%) Radiation only 2371 (16. 2%) 691 (16. 4%) Combination of Surgery /hormone /radiation 2746 (18. 8%) 614 (14. 6%)* Treatment *P<0. 05

Comparison: White vs. AA White AA 1 (highest) 6499 (44. 6%) 568 (13. 5%)* 2 3667 (25. 1%) 719 (17. 1%)* 3 2464 (16. 9%) 897 (21. 3%)* 4 1383 (9. 5%) 1072 (25. 4%)* 5 (lowest) 566 (3. 9%) 955 (22. 7%)* SES *P<0. 05

Comparison: White vs. AA White AA no 211 (1. 5%) 195 (4. 6%)* Insurance NOS 806 (5. 5%) 144 (3. 4%)* Managed care, HMO PPO 7977 (54. 7%) 2329 (55. 3%) Medicare 2781 (19. 1%) 439 (10. 4%)* Military, Veterinarian, PHS 290 (2. 0%) 251 (6. 0%)* County funded 2514 (17. 2%) 853 (20. 3%)* Insurance

Specific Aim 1 ¢ Compare mortality rates between African-Americans and Caucasians with newly diagnosed prostate cancer in Los Angeles County

Prostate-specific P=0. 002 Overall Mortality P<0. 001

Specific Aim 2 ¢ Compare the mortality rates for prostate cancer between NCI-designated Cancer Centers and other treatment facilities in Los Angeles County

Prostate-specific P<0. 001 Overall mortality P<0. 001

Specific Aim 3 ¢ Compare the mortality rates by race for patients with prostate cancer receiving care within NCI-designated Cancer Centers

Prostate-specific P=0. 52 Overall Mortality P=0. 17

Multivariate analysis • Aim 1: Compare mortality rates between African. Americans and Caucasians with newly diagnosed prostate cancer in Los Angeles County after controlling for age, SES, stage, grade, insurance, and treatment modality.

Overall mortality: AA vs. White

Prostate-specific mortality: AA vs. White

Multivariate analysis • Aim 2: Compare mortality rates for prostate cancer between NCI designated comprehensive cancer centers and other treatment facilities in Los Angeles County.

Overall mortality: NCI vs. non-NCI

Prostate-specific mortality: NCI vs. non-NCI

Multivariate analysis Aim 3: Compare mortality rates by race for patients with prostate cancer receiving care within NCIdesignated Cancer Centers ¢ Due to the small number of AA receiving care at NCI-designated cancer centers, we were unable to perform this analysis.

Specific Aim 4 ¢ Aim 4: Understand the role of sociodemographic factors in accessing care at the NCI-designated treatment centers versus non-NCI centers

Utilization of NCI-designated Cancer Center Non-NCI OR Race White 12933 (88. 7%) 1646 (11. 3%) 1. 00 AA 148 (3. 5%) 4063 (96. 5%) 0. 29 (0. 24 -0. 34)

Utilization of NCI Cancer Centers: AA vs. White

Utilization of NCI-designated Cancer Center OR (95% CI) AA vs. White 0. 42 (0. 35 -0. 50) Age at diagnosis (years) 0. 95 (0. 94 -0. 95) SES =1 highest SES =2 0. 61 (0. 53 -0. 69) SES =3 0. 41 (0. 35 -0. 49) SES =4 0. 37 (0. 30 -0. 46) SES =5 lowest 0. 30 (0. 22 -0. 42) Insurance: no 1. 00 Insurance NOS 2. 55 (1. 58 -4. 12) Managed care HMO PPO 1. 05 (0. 67 -1. 66) Medicare 3. 22 (2. 01 -5. 13) County 0. 08 (0. 02 -0. 34) Military/veteran/ PHS 2. 12 (1. 32 -3. 40)

Utilization of NCI-designated Cancer Center OR (95% CI) Grade: well-differentiated 1. 00 Moderately differentiated 3. 11 (1. 87 -5. 18) Poor/undifferentiated 3. 79 (2. 25 -6. 36) Stage: localized 1. 00 Regional 1. 29 (1. 10 -1. 51) Distant 1. 12 (0. 75 -1. 65) Treatment: watchful waiting 1. 00 Surgery only 2. 65 (2. 16 -3. 25) Hormone only 0. 80 (0. 59 -1. 09) Radiation only 1. 04 (0. 81 -1. 33) Multiple 1. 36 (1. 09 -1. 70) Year of Dx (more recent years) 1. 06 (1. 03 -1. 10)

Conclusion ¢ AA have a higher overall and prostate-specific mortality on univariate analysis l ¢ NCI-designated cancer centers have lower overall and prostate specific mortality compared to non-NCI treatment facilities l ¢ This difference persists after adjustment for all clinical and sociodemographic factors Within NCI-designated cancer centers, AA have comparable overall and prostate-specific mortality to Whites l ¢ AA have comparable overall and prostate specific mortality to Whites after adjustment for sociodemographic factors, tumor characteristic, treatment modality, and treatment site Within the constraints of the limited sample size AA are less likely to use NCI-designated Cancer Centers l Independent of SES, insurance, and tumor factors

Future Plans (Year 02) ¢ ¢ Current data set demonstrates that only 148 AA utilized the 3 NCI-designated Cancer Centers in LAC Expand the scope of analysis l ¢ Obtain data from CSP for 1976 to 2003 Explore the reasons of inferior outcomes at non-NCI designated Cancer Centers l Data from Office of Statewide Health Planning and Development. • Secondary quality indicators • Teaching status, bed size, hospital in-patient average stay, MSSA poverty, racial distribution,

Thank you!

Does Treatment Site really make a difference? ¢ In-hospital short-term mortality after Prostatectomy l High volume of prostectomies associated with low mortality • Medicare claims data • n=101, 604 between 1991 and 1994 • Nationwide Inpatient Sample • n=66, 693 between 1989 -1995 Ellison, L. M. , J. A. Heaney, and J. D. Birkmeyer, The effect of hospital volume on mortality and resource use after radical prostatectomy. J Urol, 2000. 163(3): p. 867 -9. Yao, S. L. and G. Lu-Yao, Population-based study of relationships between hospital volume of prostatectomies, patient outcomes, and length of hospital stay. J Natl Cancer Inst, 1999. 91(22): p. 1950 -6.

Does NCI designation exert an effect on outcomes ? ¢ National Cancer Act l Establish regional centers of excellence in research and patient care. l To be NCI designated • Excellence in Research • Excellence in Cancer Prevention • Excellence in Clinical Services

NCI-Designation ¢ Medicare database l Mortality after cystectomy, colectomy, pulmonary resections, pancreatic resection, gastrectomy and esophagectomy l NCI Centers had lower operative mortality in 4/6 procedures l Long term mortality: no difference Birkmeyer, N. J. , et al. , Do cancer centers designated by the National Cancer Institute have better surgical outcomes? Cancer, 2005. 103(3): p. 435 -41.

Overall HR Prostate-specific HR AA vs. White 1. 08 (0. 99 -1. 17) 1. 12 (0. 93 -1. 35) NCI center vs. non-NCI 0. 75 (0. 64 -0. 88) 0. 69 (0. 49 -0. 97) Age at diagnosis (years) 1. 05 (1. 04 -1. 06) 1. 02 (1. 01 -1. 03) SES =1 highest 1. 00 SES =2 1. 21 (1. 10 -1. 33) 1. 08 (0. 87 -1. 33) SES =3 1. 33 (1. 21 -1. 47) 1. 26 (1. 02 -1. 55) SES =4 1. 37 (1. 23 -1. 52) 1. 15 (0. 91 -1. 46) SES =5 lowest 1. 46 (1. 29 -1. 66) 1. 29 (0. 99 -1. 68) No Insurance: 1. 00 Insurance NOS 0. 53 (0. 39 -0. 71) 0. 46 (0. 25 -0. 84) HMO PPO 0. 84 (0. 67 -1. 05) 0. 91 (0. 60 -1. 39) Medicare 0. 86 (0. 68 -1. 08) 0. 82 (0. 53 -1. 28) Military/Vet/ Indian/PHS 0. 80 (0. 61 -1. 06) 0. 70 (0. 40 -1. 22) County funded 0. 82 (0. 65 -1. 03) 0. 74 (0. 48 -1. 15)

Overall HR Prostate-specific HR 1. 00 Moderately differentiated 1. 08 (0. 91 -1. 28) 1. 68 (0. 91 -3. 07) Poor/undifferentiated 1. 69 (1. 42 -2. 01) 4. 87 (2. 66 -8. 92) 1. 00 Regional 1. 25 (1. 07 -1. 38) 2. 28 (1. 76 -2. 94) Distant 3. 76 (3. 38 -4. 18) 14. 72 (12. 21 -17. 76) Grade: well-differentiated Stage: localized Treatment: watchful waiting 1. 00 Surgery only 0. 31 (0. 28 -0. 35) 0. 19 (0. 13 -0. 27) Hormone only 1. 04 (0. 95 -1. 15) 1. 29 (1. 05 -1. 60) Radiation only 0. 61 (0. 55 -0. 68) 0. 50 (0. 36 -0. 70) Multiple 0. 67 (0. 60 -0. 74) 0. 86 (0. 67 -1. 09) Year of Dx 1. 03 (1. 01 -1. 06) 0. 98 (0. 93 -1. 03)