Download Obesity and prostate cancer incidence and survival Elizabeth

Obesity and prostate cancer
incidence and survival
Elizabeth A. Platz, ScD, MPH
Professor and Martin D. Abeloff, MD Scholar in Cancer Prevention
Department of Epidemiology,
Johns Hopkins Bloomberg School of Public Health;
Department of Urology and the James Buchanan Brady Urological Institute,
Johns Hopkins School of Medicine;
Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins
Body fatness and prostate cancer
incidence and mortality
2007
2007
Methodologic issues that arise as a result
of routine PSA-based screening
for prostate cancer
• Changing spectrum of disease due to PSA screening
– Proportion of advanced vs early disease
– Nature of the early cases (may never have become
clinically apparent)
– Use of PSA screening differs around the world
• Detection bias
– Differential opportunity to be PSA screened by exposure
status (obese vs not obese)
– Differential detection of the presence of disease due to
exposure changing PSA levels (lower PSA in obese men)
Influence of PSA screening
on the observed association
between obesity and prostate cancer risk
Renehan AG et al Eur J Cancer 2010;46:2581-92.
Death from prostate cancer by BMI,
Cancer Prevention Study II, 1982-1998
RR
1.4
P-trend < 0.001
1.3
1.2
1.1
1
18.5-24.9
25.0-29.9
30.0-34.9
BMI (kg/m2)
Calle E et al. 2003; NEJM 348:1625-1638
35.0+
Obesity and prostate cancer,
Health Professionals Follow-up Study
RR
1.9
1.8
1.7
1.6
1.5
1.4
1.3
1.2
1.1
1
P-trend=0.47
P-trend=0.03
RR=1.80, 95% CI 1.10-2.93
combined
Fatal
< 21
21-22.9
BMI kg/m2
23-24.9
25-27.4
Incident
27.5-29.9
Giovannucci E et al. Int J Cancer 2007;121:1571-8. PMID: 17450530
>= 30
Obesity and prostate cancer risk in the
NIH-AARP Diet and Health Study
RR
2.12
1.08-4.15
2.3
2.1
P-trend = 0.02
1.9
1.7
1.5
1.3
1.1
0.9
< 25
25-29.9
0.7
30-34.9
>= 40
BMI (kg/m2)
0.5
Incident
Wright ME et al. Cancer. 2007;109:675-84
35-39.9
Fatal
0.67
0.50-0.89
P-trend = 0.0006
Obesity and prostate cancer,
Melbourne Collaborative Cohort
RR
1.7
Per 5 kg/m2:
RR=1.49, 95% CI 1.11-2.00
P-trend=0.01
RR=1.52,
95% CI 0.89-2.58
1.5
Per 5 kg/m2:
RR=0.99, 95% CI 0.89-1.10
P-trend=0.83
1.3
1.1
0.9
Ref
Ref
Fatal
Nonaggressive
0.7
0.5
BMI kg/m2
<23
23-24.9
25-29.9
>= 30
Bassett JK et al. Int J Cancer 2012;131:1711-9. PMID: 22213024.
Weight gain since age 18 and prostate
cancer, Melbourne Collaborative Cohort
RR
2.1
Per 5 kg:
RR=1.13, 95% CI 1.02-1.26
P-trend=0.02
RR=1.84,
95% CI 1.09-3.09
1.9
Per 5 kg:
RR=0.99, 95% CI 0.81-1.22
P-trend=0.77
1.7
1.5
1.3
Ref
Ref
1.1
0.9
Fatal
Nonaggressive
0.7
0.5
Weight gain since age 18 kg
<5
5-9.9
10-19.9
>=20
Bassett JK et al. Int J Cancer 2012;131:1711-9. PMID: 22213024.
BMI and prostate cancer by grade,
Prostate Cancer Prevention Trial
ORmv
1.4
P-trend=0.03
P-trend=0.04
1.2
BMI
1
Q1
Q2
Q3
Q4
0.8
0.6
0.4
0.2
0
Total
Low grade
High grade
N=1,936
N=1,300
N=521
Gong Z et al. Cancer Epidemiol Biomarkers Prev 2006;15:1977-1983
Meta-analyses
•
•
•
•
•
•
•
Bergstrom A, Pisani P, Tenet V, Wolk A, Adami HO. Overweight as an avoidable
cause of cancer in Europe. Int J Cancer 2001;91:421–30
MacInnis RJ, English DR. Body size and composition and prostate cancer risk:
systematic review and meta-regression analysis. Cancer Causes Control
2006;17:989–1003
Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and
incidence of cancer: a systematic review and meta-analysis of prospective
observational studies. Lancet 2008;371:569–78
Cao Y, Ma J. Body mass index, prostate cancer-specific mortality, and biochemical
recurrence: a systematic review and meta-analysis. Cancer Prev Res 2011;4:486501
Discacciati A, Orsini N, Wolk A. Body mass index and incidence of localized and
advanced prostate cancer--a dose-response meta-analysis of prospective studies.
Ann Oncol 2012;23:1665-71
Allott EH, Masko EM, Freedland SJ. Obesity and prostate cancer: weighing the
evidence. Eur Urol 2013;63:800-9
Golabek T, Bukowczan J, Chłosta P, Powroźnik J, Dobruch J, Borówka A. Obesity
and Prostate Cancer Incidence and Mortality: A Systematic Review of Prospective
Cohort Studies. Urol Int. 2013 Aug 8. [Epub ahead of print].
“Results: The evidence from the prospective cohort studies linking obesity
with PCa incidence has not been consistent. However, cumulative data is
compelling for a strong positive association between obesity and fatal PCa.”
Meta-analysis:
BMI and prostate
cancer risk, by
localized and
advanced disease
Literature search to Oct 2011.
Discacciati A et al. Ann Oncol 2012;23:1665-71.
Meta-analysis:
BMI and fatal prostate cancer
per 5 kg/m2 increase in BMI
Cao Y and Ma J. Cancer Prev Res 2011;4:486-501. PMID: 21233290.
Cause versus bias?
Hypotheses based on published results
• Obesity 
– Risk (causation)
•  Risk of death from prostate cancer
•  Risk of nonaggressive prostate cancer
– Detection bias
•  PSA (lower production, hemodilution)
•  Prostate volume
Freedland SJ, Platz EA Epidemiol Rev 2007;29:88-97;
Freedland SJ et al. Cancer Causes Control 2006;17:5-9;
Freedland SJ et al. J Urol 2006;175:500-504;
Bañez LL et al. JAMA 2007;298:2275-2280
If causal, what pathways mediate the
obesity-prostate cancer association?
• Metabolic and hormonal perturbations secondary
to obesity?
– Insulin regulation (and IGF-axis)
– Sex steroid hormones
– Others?
• Bioactive factors secreted by adipocytes?
– Energy regulation
– Inflammatory mediators
– Others?
• Obesity’s influence on the detectability of prostate
cancer via these same pathways?
Is obesity associated with poor outcome
after prostate cancer diagnosis?
• Emerging support that the extent of body
fatness and weight gain
– Before diagnosis
– Circa diagnosis
• are risk factors for recurrence and prostate
cancer death in men with prostate cancer.
• Excellent review: Cao Y and Ma J. Cancer Prev Res
2011;4:486-501. PMID: 21233290.
Meta-analysis:
BMI and death from prostate cancer
in men with the diagnosis
per 5 kg/m2 increase in BMI
Cao Y and Ma J. Cancer Prev Res 2011;4:486-501. PMID: 21233290.
Meta-analysis: BMI and prostate
cancer recurrence after treatment
per 5 kg/m2 increase in BMI
Cao Y and Ma J. Cancer Prev Res 2011;4:486-501. PMID: 21233290.
Literature is not perfectly consistent
for obesity and prostate cancer
recurrence
• Publications since Cao and Ma’s meta-analysis
– No association between obesity and recurrence
• Lee SE et al. BJU Int 2011;107:1250-5. PMID: 20880194. [Korea]
• Narita S et al. Prostate Cancer Prostatic Dis 2013;16:271-6. PMID:
23752230. [Japan]
• Tomaszewski JJ et al. Urology 2013;81:992-6. PMID: 23453649.
– Positive association between obesity with recurrence
• Joshu CE et al. Cancer Prev Res 2011;4:544-51. PMID:
21325564. [also weight gain]
• Asmar R et al. Prostate Cancer Prostatic Dis 2013;16:62-6. PMID:
22907512.
Men who gain weight have a higher risk
of prostate cancer recurrence after
prostatectomy, JHH
4
P for trend
0.02
3
RR
OR 2
1
0
Weight Loss
>2.2 kg
Maintenance
<2.2kg
Weight Gain
>2.2 kg
Weight change from 5 years before to 1 year after surgery
Adjusted for weight 5 years before surgery, height, physical activity 1 year after surgery, age, race/ethnicity, family history,
year of surgery, stage, grade, and smoking status.
Joshu CE et al. Cancer Prev Res 2011;4:544-51. PMID: 21325564.
Considerations (fully or partially)
specific to prostate cancer
outcomes
• To confirm or refute associations observed
thus far
• For studies going forward
Consideration 1
• What is the optimal prostate cancer outcome to
capture biology and import?
– Biology and import: Progression to metastasis or death
from prostate cancer (rather than biochemical
recurrence).
– Nature of outcome that can be studied may depend on
the type of treatment, though.
• Especially when studying men with clinically localized
prostate cancer (patients selected for curability  few
deaths)
– Surgery - biochemical recurrence
– Radiation therapy or active surveillance - rising PSA
Consideration 2
• Confounding - Body fatness appears to be a risk
factor for advanced stage and high-grade prostate
cancer AND stage and grade are prognostic factors
– Thus, must take into account stage and grade in the
analysis to determine whether body fatness/weight gain are
associated with poor outcome in men with prostate cancer
Body fatness /
weight gain
Advanced stage /
high-grade
prostate cancer
Recurrence / death
from prostate cancer
Consideration 3
• Etiologically relevant measurement - Timing of body
fatness / weight gain relative to the diagnosis /
treatment of prostate cancer
– Pre-diagnostic
– At diagnosis / treatment
– Post diagnosis / treatment
Normal
Body fatness /
weight gain
Body fatness /
weight gain
Precursors
Organ-confined
Prostate cancer
Body fatness /
weight gain
Limited
Extraprostatic
Body fatness /
weight gain
Disseminated
Fatal
How can body fatness influence
outcomes after treatment, especially
after prostatectomy?
Body fatness Body fatness Body fatness
Escaped prostate
cancer cells
Nascent
bony mets
Body fatness
No prostate,
Prostate cancer Prostatectomy
no prostate
focus
cancer focus
Body fatness
Recurrence / death
from prostate cancer
Consideration 4
• Does the influence of weight gain on outcome differ by
starting body fatness?
–
–
–
–
Lean, gain weight
Obese, gain weight
Lean, no weight gain
Obese, no weight gain
• Considerations 3 and 4 coupled: Full evaluation of body
fatness and weight gain over the life course.
– Requires prospective study of men without the diagnosis, followed
to diagnosis (and treatment), and then followed to death
– Requires repeated measures of body fatness, including circa the
diagnosis /treatment
Consideration 5
• Confounding and modifying effects by factors
that are highly correlated with body fatness
–
–
–
–
Physical inactivity
Diabetes
Energy intake
Smoking
RRobs > RRtrue
Obesity
+
Poor outcome
Physical
inactivity
+
• Statistical analyses
RRobs < RRtrue
Obesity
-
Poor outcome
Smoking
+
Consideration 6
• Alternative explanations to biology
• Greater technical difficulty when treating obese men
relative to lean men (e.g., positive surgical margins)
– Lower likelihood of cure unrelated to prostate cancer
biology
Ho T et al. Eur Urol 2012;62:910-6. PMID: 22921964
Consideration 6
• Greater technical difficulty when treating obese men
relative to lean men (e.g., positive surgical margins)
Ho T et al. Eur Urol 2012;62:910-6. PMID: 22921964
Consideration 6
• Alternative explanations
• Different choice of treatment by obese and nonobese men
– Where the treatment may have a different likelihood of cure
irrespective of extent of body fatness
Consideration 7
• Hormonal therapy for men with metastatic prostate
cancer causes central adiposity and metabolic
perturbations.
– What is the influence of this milieu on prostate cancer
death (beyond obvious increase in risk of death from other
causes)?
– An area of active study.
Consideration 8
• Surveillance for the early recurrence (e.g., postprostatectomy)
– What is the influence of body fat on PSA produced by
cells that have escaped from the prostate?
•
 PSA (lower production, hemodilution) in men who are obese
compared with lean
– Detection bias - time to recurrences would be falsely
LONGER in obese compared with lean men
• Needs thorough evaluation.
BMI and risk of biochemical recurrence
after prostatectomy, JHH
Covariates
At the time of surgery
Freedland SJ et al. J Urol 2005;174:919-22. PMID: 16093988.
Pre-diagnostic obesity and prostate
cancer death in men with prostate cancer,
Physicians’ Health Study
Unadjusted
Ma J et al. Lancet Oncol 2008;9:1039-47. PMID: 18835745.
Pre-diagnostic body mass index and
prostate cancer death in men with
prostate cancer, PHS
P-trend=0.0042
2
1.8
1.6
RR*
1.4
1.2
*Adjusted for age at diagnosis, baseline
smoking status, time interval from BMI
measurement to prostate-cancer diagnosis,
clinical stage, and Gleason grade.
BMI 30+ vs < 25 kg/m2:
Adj excluding for stage/grade – RR=2.66
Adj including for stage/grade – RR=1.95
Body fatness /
weight gain
Advanced stage /
high-grade
prostate cancer
Recurrence / death
from prostate cancer
1
< 25
25.0-29.9
BMI (kg/m2)
Ma J et al. Lancet Oncol 2008;9:1039-47. PMID: 18835745.
30.0+
Men who gain weight have a higher risk
of prostate cancer recurrence after
prostatectomy, JHH
4
P for trend
0.02
3
RR
OR 2
1
0
Weight Loss
>2.2 kg
Maintenance
<2.2kg
Weight Gain
>2.2 kg
Weight change from 5 years before to 1 year after surgery
Adjusted for weight 5 years before surgery, height, physical activity 1 year after surgery, age, race/ethnicity, family history,
year of surgery, stage, grade, and smoking status.
Joshu CE et al. Cancer Prev Res 2011;4:544-51. PMID: 21325564.
Summary
• Evidence is mostly consistent that obesity is a risk
factor for a more aggressive prostate cancer
phenotype (e.g., high grade/advanced stage
disease, prostate cancer mortality).
• Evidence building that obesity/weight gain is a risk
factor for poor outcome in men diagnosed with
prostate cancer.
• Many methodologic issues still need to be addressed
for both etiology and prognosis.
Important questions that remain to
be addressed
• Does weight loss reduce the risk of developing
aggressive prostate cancer?
Men who gain weight have a higher risk
of recurrence after prostatectomy, JHH
• Does weight loss
reduce the risk of
recurrence, the
development of
metastases, and death
from prostate in men
with the disease?
Joshu CE et al. Cancer Prev Res 2011;4:544-51. PMID: 21325564.
Take home message for men
• Maintaining a healthy weight is important
for good health in general, and may help
prevent dying from prostate cancer.