California Cancer Research Program
Completed Projects
The following Principal
Investigators have completed their research as funded by CRP. Where applicable,
publications resulting from CRP funding are also included.
CYCLE I FINAL LAY
ABSTRACTS
Epidemiology
Investigator-Initiated
Awards
Victoria K. Cortessis, Ph.D.: Genetic Epidemiology of Testicular Cancer
and Cryptorchidism
Paul
B. English, Ph.D., M.P.H.: Exploring Perinatal Risk Factors for
Testicular Cancer
Sue Ann Ingles, Dr.P.H., M.A.: Advanced Prostate Cancer-Risk Factors in
African-Americans
Esther M. John, Ph.D.: Lifestyle Factors and Risk of Advanced Prostate
Cancer
Allan H. Smith, M.D., Ph.D.: California Arsenic Methylation Study
Alice S. Whittemore, Ph.D., M.A.: Prostate Cancer Survival in a
Multi-ethnic Cohort
Pilot and Feasibility
Study Awards
James J. Beaumont, Ph.D., M.S.P.H.: Bias from Loss to Interview in
Prostate Cancer
Epidemiology
Jan M. Van Tornout, M.D., M.S.:
Steroid Receptor Genes and EwingÕs
Sarcoma of Bone in Children
Population-Based
Epidemiology Targeted RFAs
Sharan L. Campleman, Ph.D., M.P.H.: Prostate Cancer in Rural California:
Diagnosis and Treatment
Rosemary D. Cress, Dr.P.H.: Quality of Care
for Women with Ovarian
Cancer
Paul K. Mills, Ph.D.: Prostate and Testicular Cancer in Farmworkers
Paul K. Mills, Ph.D.: Triazine Herbicides and Ovarian Cancer Risk
ABSTRACTS
Bias from Loss to
Interview in Prostate Cancer Epidemiology
James J. Beaumont, Ph.D., M.S.P.H.
University of California,
Davis
Pilot and Feasibility
Study Award, $122,104.00 / 24 mos.
(Prostate Cancer)
Prostate cancer cases lost to follow-up due to death,
changing residence or incomplete information can cause bias in case-control
studies. This study quantifies this loss in the California Cancer Registry by
the length of time from diagnosis, reporting time to the registry, geographical
area and site of diagnosis. Study subjects were cancers diagnosed in 1996 and
1997 from three regional registries in agricultural areas of California (Region
2, Central; Region 3, Sacramento; and Region 6, North) and reported to the
California Cancer Registry by May 2000. We calculated the time between
diagnosis and report to the registry for the 60,868 cases of all types, and we
randomly selected 2000 cases of prostate cancer and 6000 cases of other types
of cancer. Our objective was to identify cases that were interviewable, which
we defined as alive and residing at the same residence as at diagnosis. Based
on all 60,868 cases, the median time between diagnosis and reporting to the
registry for all types of cancer was 373 days, and it differed with respect to
type of cancer, geographical region and race/ethnicity. In our sample of 8,000,
and using only registry data for follow-up, we estimated that 66% of the cases
were interviewable three months after diagnosis, 36% were interviewable after 1
year and 15% after two years. We found differences in the natural reporting
time to the CCR by type of cancer, geographical region and race/ethnicity. We
also showed that interviewablity decreases with time from diagnosis, thus
loss-to-interview could affect risk estimates in registry based case-control
studies.
Sharan L. Campleman,
Ph.D., M.P.H.
Public Health Institute
Population-Based Epidemiology Targeted RFA, $266,715.00 / 36
mos.
(Prostate Cancer)
Although
diagnosis of prostate cancer has increased substantially in recent years,
considerable medical debate continues to focus on the value and best
application of early cancer screening, and, age- or disease- appropriate treatment
strategies. Such controversies in
the therapeutic management of prostate cancer may lead to variations in the
current practices of early screening and applied therapies, especially in older
men with localized disease. The objectives of this study are to determine the
factors likely to influence diagnostic and treatment choices for prostate
cancer in rural Northern California.
Rural populations, which are often characterized by levels of higher
poverty, lower education, fewer health care facilities and medical specialties,
may be an area particularly affected by physical and economic isolation.
Primary,
malignant prostate cancer cases initially diagnosed between 1995 and 1997
(N=2,935) among the residents of the sixteen northern-most counties monitored
by the Cancer Registry of Northern California (CRNC), Region 6 of CCR were
eligible for the study (reported as of November 1999). The CRNC prostate cancer database was
enhanced through additional medical record review in order to obtain information
on comorbidities, vital status, and treatment confirmation, excluding pathology
reports, autopsy, and death certificate only cases (n=108) or cases with
medical charts located outside regional facilities/physicians (n=467). The more rural counties of the
region (< 50% urban population) had substantially lower overall AAIR,
103.2/100,000 then the urban counties (> 50 % urban population),
113.0/100,000, in Region 6. Local
stage prostate cancer incidence was significantly lower statistically in the
rural (46.5/100,000) vs. urban (59.6/100,000) counties. Differences appear greatest among the
oldest cases (75 years of age or older) – local stage 179.2 vs. 291.7
(significant p<0.05), regional/distant stage 62.7 vs. 124.9 (significant
p<0.05), unknown/missing stage 469.8 vs. 380.8 (not significant), rural vs.
urban counties, respectively.
Multivariate analysis
included age, comorbidity, race, tumor grade; hospital type, rural residence
and neighborhood income identified several factors significantly associated
with early diagnosis and treatment variation. Men residing in Northern
California were more likely to be diagnosed with localized versus advanced
disease if under age 65 and residing in a more urban area. Unstaged cases were highly associated
with older age, lower grade, lower income and rural residence. Surgical
treatment for localized prostate cancer was significantly associated with
younger age, lower comorbidity, lower income and rural residence, in comparison
with cancer-directed radiation or hormone therapy that was significantly
associated with older age, lower comorbidity, higher income, and urban
residence. Relative to non-Hispanic whites, men from other race/ethnicities
were less likely to undergo more aggressive therapies such as surgery (whether
radical prostatectomy or local tumor destruction) and external beam radiation,
even after accounting for income and rural residence. Additionally, men not undergoing any treatment as a planned
first course of cancer therapy for localized prostate cancer were significantly
more likely to be older (over 75 years), to have more comorbidities, to live in
a rural area, and to be identified as any race other than non-Hispanic white.
Genetic Epidemiology of
Testicular Cancer and Cryptorchidism
University of Southern
California
Investigator-Initiated
Award, $1,262,719.00 / 36 mos.
(Testicular Cancer)
Background:
Testicular cancer affects young men, primarily those 15 to 49 years of
age, and the incidence of this cancer has been rising steadily for over a
century. Until recently,
testicular cancer was the leading cancer among men aged 15-29, and it is now
second only to the AIDS-related cancer, KaposiÕs sarcoma. Advances in treatment dramatically
improved survival during the last 25 years. Unfortunately, during the same interval, incidence rates
again doubled, and it became clear that even successfully treated testicular
cancer often has serious complications including infertility, sexual
dysfunction, elevated risk of cancer in the remaining testicle, and elevated
risk of other cancers. There is a
strong consensus that the rapid and steady rise in incidence can be explained
only by increasing exposure to one or more environmental causes. By identifying causes of testis cancer,
we hope to understand why it is becoming more common, and to develop preventive
measures that may reverse this trend.
Unfortunately, current
insights do not suggest preventive measures. Established risk factors for testis cancer are a family
history of testis cancer, and a personal history of undescended testicles, or
cryptorchidism, a congenital condition in which baby boys are born with one or
both testicles in the abdomen, rather than in the scrotum. Cryptorchidism may predispose boys to
later develop testicular cancer either because the abdominal location of a testicle increases the risk of
testicular cancer or because both
conditions (undescended testicles and testicular cancer) have one or more
causes in common. Other research
findings suggest that events occurring in utero may also predispose to testis
cancer. Familial occurrence of
testis cancer suggests inherited genetic factors are also important, but
specific genes have not been identified.
Although testicular cancer appears to have both genetic and
environmental causes, we began by searching for genetic causes, because they
may be easier to find, since inherited genetic constitution is relatively
stable. In contrast,
important environmental exposures may occur early in life and can change over
time, interfering with their accurate measurement after cancer is
diagnosed. We hope that once
important genes are identified, the biological function of their products will
suggest environmental causes.
Results: This research was designed to find genes that predispose to testis cancer. Two complementary approaches can reveal genes causing human disease: (1) genetic linkage studies conducted among families with two or more affected members, and (2) genetic association studies comparing affected individuals with others. We began this research with a genetic linkage study (1), but in the process of conducting it we are able to identify, at essentially no additional cost, populations that may provide unique, complementary insights through genetic association studies. Over 3000 men participated in the screening phase of the research that identified over 300 multiple case families qualifying for the genetic linkage study; and members of the majority of these families chose to participate. At the time of submission of this report, analyses were not complete enough to reach meaningful conclusions, but power calculations show that these families should provide enough information to localize a single major gene predisposing to testicular cancer and cryptorchidism if there is only one such gene, or a substantial amount of the required information if there are multiple genes of this kind. In 2003 our research team hosted scientific and planning meetings with other prominent researchers, which resulted in our joining in an international collaborative effort to use linkage analysis to identify genes that predispose to these conditions. We submitted a major proposal to the National Cancer Institute to continue and expand the linkage study (Cortessis, PI) and await its outcome. To prepare for association studies, we also identified over 500 men who seem more likely to have a genetic etiology because of a personal history of either bilateral testis cancer (146), or of both testis cancer and cryptorchidism (372). We received two grants to conduct association studies (2) of specific candidate genes in these populations: an award from the Robert E. and May R. Wright Foundation (Cortessis, PI) to study bilateral testis cancer, and a cycle III award from the CRP (Cortessis, PI) to study testis cancer with cryptorchidism. With additional funds from the STOP!CANCER foundation (Cortessis, PI) we conducted a pilot study of cryptorchidism in collaboration with physicians of the Urology Division at Childrens Hospital Los Angeles (CHLA); this study produced pilot data to be used in a proposal to study candidate genes for pediatric cryptorchidism, which we plan to submit to the National Institutes of Health in the fall of 2004. Dr. Cortessis has presented rationale for candidate genes, preliminary findings, and/or a postulated model for the etiology of these conditions at USC, UCLA, The City of Hope, the Southwest Oncology Group and the Society for Epidemiologic Research annual meeting. Preliminary analyses of the screening data assembled with this award and data describing the boys enrolled at CHLA suggest that familial aggregation of testis cancer and cryptorchidism is greater than previously reported, supporting the hypothesis of genetic causes. We will provide the CRP with manuscripts detailing more conclusive analyses as they are prepared.
Quality of Care for Women
with Ovarian Cancer
Public Health Institute
Population-Based
Epidemiology Targeted RFA, $327,424.00 / 36 mos.
(Ovarian Cancer)
The purpose of this research project was to evaluate care of women with ovarian cancer diagnosed in Northern California. The study population consisted of 2,150 women residing in Northern California with a first diagnosis of primary epithelial ovarian cancer between January 1994 and December 1996. Cases were identified through the California Cancer Registry, and their physicians were surveyed to supplement registry treatment information. Almost 89% of women under age 75 with FIGO Stage III or IV tumors received chemotherapy, with levels of treatment highest for women diagnosed at Stage III. Patients aged 75 or over were significantly less likely than younger women to receive adjuvant chemotherapy (58.2% vs. 86.1%, p=0.001) regardless of stage at diagnosis. Approximately 20% of patients under age 55 with early stage (Stage IC and II) cancer received no chemotherapy. Treatment in an American College of Surgeons hospital and treatment by a gynecologic oncologist increased the likelihood of receiving chemotherapy. Hospitalization for comorbid illness, race/ethnicity, census-based measures of socioeconomic status, size or teaching status of hospital were all unrelated to probability of treatment after adjustment for other factors. Reasons reported most frequently by physicians for no treatment were lack of clinical indication and patient refusal. The results of this study suggest that, despite scientific evidence and published guidelines that advocate adjuvant chemotherapy for most women with ovarian cancer, some groups of women did not receive optimum treatment. Analysis of the survival experience of this patient cohort is ongoing.
Paul B. English, Ph.D., M.P.H.
Impact Assessment, Inc.
Investigator-Initiated
Award, $751,628.00 / 36 mos.
(Testicular Cancer)
Testicular cancer is the most
common cancer among young men in the U.S., and the rate of new cases of this
cancer has increased 51% between 1973 and 1995. Due to the early age of diagnosis and results from
scientific research, it is thought that hormone exposures (both natural and
synthetic) around the time of birth may play a role in the development of this
illness. The goals of this project
were to (1) examine the association between risk factors which are markers of
estrogen exposure in the mother at the time of birth and risk of testicular
cancer in their male offspring; (2) Examine the association between population
socio-demographic factors and the risk of testicular cancer; and (3) Examine
the association between testicular cancer risk and population exposures which
may be cancer promotors (such as drinking water contaminants and agricultural
pesticide use) based on the address at diagnosis.
Confirming previous studies,
we found that men who were born to older mothers, and men who were first or
second born in their families, had higher risks of testicular cancer than men
who were born to younger mothers or than men who had older siblings. Laboratory studies have analyzed blood
from pregnant women and have found that older women and women giving birth to
their first baby have higher estrogen levels than younger women and women who
have delivered previously. These
findings confirm a possible role of high estrogen levels in the mother in
causing testicular cancer in their male offspring. This relationship between elevated estrogen levels and
cancer has also been suggested for breast and endometrial cancers. The relationship between older age in
the mother and increased testicular cancer risk in their sons has been
suggested to be due also to the transfer of damaged DNA from mother to son.
We also found that among male
infants who are born on time (greater than 37 weeks of pregnancy), those that
are born small (less than 5.5 pounds) have more than twice the risk of
testicular cancer than those of normal weight (greater than 5.5 pounds). We found this effect to be primarily in
one type of testicular cancer, the seminomas, which tend to occur in men older
than 20. This finding supports previous reports that normal development of the
fetus late in pregancy lowers the risk of testicular cancer and suggests that
interventions to increase birthweight could be helpful.
We have completed an analysis
of pesticide use near residences in adulthood and testicular cancer risk and
found elevated risks for living near agricultural fields which have been
applied with compounds which are endocrine disruptors, possible carcinogens,
genotoxics, and reproductive toxics, but found no dose-response patterns. Since we did not measure individual
exposure to pesticides and did not measure pesticide exposures near the time of
birth, these results are only suggestive of an association and should not be
interpreted as a definitive link between pesticide exposure and cancer
risk. Analysis of cruder data on
historic pesticide use or crop patterns near the time of birth did not reveal
any meaninful associations. Linkage
of testicular cancer cases and control addresses to water sources allowed us to
analyze mean levels of water contaminants delivered to households (up to 2
years before diagnosis), adjusting for multiple water sources and
treatments. The most interesting
finding was that of a synergistic effect of nitrates and trihalomethanes in
producing a dose-reponse effect on testicular cancer risk. Since we lacked water consumption data
and were unable to assess exposure further back than two years before
diagnosis, this finding should be interpreted with caution.
Advanced Prostate Cancer -
Risk Factors in African-Americans
University of Southern
California
Investigator-Initiated
Award, $1,499,943.00 / 36 mos.
(Prostate Cancer)
Our major objective was to
identify 400 African-American men living in Los Angeles County who have been
diagnosed with advanced prostate cancer, and to conduct interviews and collect
blood samples from these men and from matched controls (men without prostate
cancer), and to extract DNA from the blood samples and perform genetic testing
for a number of candidate genes.
At the end of this study, we
have enrolled 381 men with advanced prostate cancer and 133 controls (men
without prostate cancer) in Los Angeles County. Along with our sister study in Northern California (CCRP 99-00527V-10182,
PI: John), we now have blood samples on a total of 488 African-American cases
(381 from southern + 101 from northern California) and 218 African-American
controls (133 from southern and 85 from northern California).
Based on the success of our
recruitment efforts in this study, we received additional funding from the
National Institutes of Health (NIH) to expand this study and extend recruitment
to whites and Hispanics in Los Angeles County.
Although complete results of this study are still pending completion of the expanded study, we have generated preliminary evidence suggesting that among African-Americans, both androgens and vitamin D are important in prostate cancer (with androgens promoter cancer development and vitamin D protecting against cancer development). African-American men are more likely to carry genetic variants that confer higher androgen activity and higher PSA levels. In addition, African-American men are more likely to be deficient in vitamin D. Among control men in this study, 40% of African-Americans, but only 9% of whites were deficient in vitamin D. Further results will be published upon completion of the expanded study funded in part by the NIH.
Lifestyle Factors and Risk
of Advanced Prostate Cancer
Northern California Cancer
Center
Investigator-Initiated
Award, $1,002,506.00 / 36 mos.
(Prostate Cancer)
Objectives: We
conducted a case-control study in White and African-American men living in the
San Francisco Bay Area to examine the effect of various lifestyle factors on
the risk of developing advanced prostate cancer. We assessed whether (1) high
levels of exposure to vitamin D (from sunlight exposure and dietary vitamin D
intake) and high levels of physical activity decrease risk, and (2) high
dietary calcium intake and obesity increase risk. Thus, the focus of this study
was on lifestyle factors that are potentially modifiable. The study also
collected blood or mouthwash samples from participants in order to assess
whether certain common alterations in the vitamin D receptor gene are
associated with the risk of developing advanced prostate cancer. Frozen DNA,
plasma and serum were stored for future studies of advanced prostate cancer.
Methods:
Men newly diagnosed with advanced prostate cancer between 1998 and 2000
(cases) were identified through the regional cancer registry. Men who have
never been diagnosed with prostate cancer served as the comparison group
(controls) and were identified through random-digit dialing and through the
rosters of the Health Care and Financing Administration (HCFA). Cases and
controls were invited to participate in a brief screening telephone interview
to determine study eligibility, to complete an in-person interview and
measurements (weight, height, waist and hip circumferences, skin pigmentation,
resting pulse rate) usually conducted at the participantÕs home, and to provide
a blood or mouthwash sample. The interview data from this study were combined
with data for cases diagnosed between 1997 and 1998 that were collected in a
CRP companion study using the same study protocol and questionnaire. The
statistical analysis of the combined data was based on 568 cases (450 Whites,
118 African-Americans) and 545 controls (455 Whites, 90 African-Americans).
Biospecimens were collected for 533 cases (426 Whites, 107 African-Americans)
and 525 controls (440 Whites, 85 African-Americans).
Study findings: We
assessed lifetime activity from exercise, transportation (walking, bicycling),
strenuous household and outdoor chores, and occupation, and found that men with
more vigorous activity had a 23% lower risk of advanced prostate cancer.
Moderate activity did not reduce risk. No clear findings emerged for height,
weight, and body mass index. Large hip and waist sizes increased risk by
30-60%. We had hypothesized that men with more sun exposure would have a lower
risk of advanced prostate cancer, and found risk reductions ranging from
20-30%. High outdoor activity reduced risk among African-Americans, whereas
measured sun exposure (i.e., the difference in skin pigmentation between the
upper inner arm and forehead) reduced risk among Whites. The production of
vitamin D following sun exposure is known to be higher in individuals with
lighter skin. Among African-Americans, risk was reduced among those with
lighter skin. No association was found among Whites. Dietary vitamin D intake
was not associated with risk, whereas high calcium intake increased risk.
Several common alterations in the vitamin D receptor gene increased risk by
35-45%, and it appears that men with certain alterations in this gene may have
a greater benefit from sun exposure.
Conclusions: Our
findings support the importance of modifiable lifestyle factors in reducing the
risk of advanced prostate cancer, and warrant further research on the role of
physical activity and vitamin D in the development of this cancer.
Prostate and Testicular
Cancer in Farm Workers
Public Health Institute
Population-Based
Epidemiology Targeted RFA, $410,469.00 / 36 mos.
(Prostate Cancer)
The objectives of this study
were to evaluate risk of two male cancers, prostate and testis cancer, in the
membership of a large agricultural farm worker union, the United Farm Workers
of America (UFW). In order to accomplish these objectives the resources of the
California Cancer Registry (CCR) were utilized in order to identify all newly
diagnosed prostate and testis cancers in the UFW from 1988 through 1999.
The first step in this study
was to construct a roster of al past and current members of the UFW. This was
done by compiling a list from two of the benefits programs offered to all
members of the union, namely the Robert F. Kennedy Farm Workers Medical Plan
and the Juan De LA Cruz Pension Program. The initial electronic roster included
identifying information on 146,000 "ever" members of the union.
However, the quality of these data was found to be questionable and a number of
validity checks were completed to improve the validity of the data.
Specifically, a computer program (SSNChecker) was used to check the validity of
the social security numbers included in the original database. The resulting
validated database included information on approximately 139,000
"ever" union members.
In order to identify newly
diagnosed cancers in this membership roster, the files of the UFW were checked
against the database of the CCR for the years 1988 through 1999 using the
INTEGRITY matching software. This matching program successfully identifies
1,001 newly diagnosed cancers in the UFW. This number included 222 prostate
cancer cases and 5 cases of testis cancer. Using a proportionate approach the
risk of prostate and testis cancers in the UFW was compared to all Hispanic
males in California. For prostate cancer the resulting Proportionate Cancer
Incidence Ratio (PCIR) was 0.93 indicating no elevation in risk. For testis
cancer, the PCIR was 0.74. Neither of these findings were statistically
significant.
The next step in this process
was to evaluate risk of prostate cancer associate with several characteristics
of the farm workers including the types of crops they commonly cultivated, the
date of first union involvement, the duration of union membership. In addition,
in order to assess the impact of specific pesticides, detailed work histories
of the farm workers were linked with the Pesticide Use reports (PUR) of the
California Department of Pesticide Regulation (DPR). For each farm worker, the
types of crops and commodities worked in, the location of work and the dates of
work were matched with the records of the DPR. This information was obtained
for each of the 220 prostate cancer cases and for five healthy
"control" workers who were similar in age to the cases. The results
of the case control study indicated that woke in certain crops (e.g. mushrooms)
was associated with slightly higher prostate cancer risk (O.R.=1.9) but his
result was not statistically significant. Risk of prostate cancer actually decreased
with increasing duration of union affiliation. However, after taking account of
age, date of first union membership and duration of membership, risk of
prostate cancer was found to be associated with use of the pesticides simazine,
Heptaclor and Lindane.
1. Mills PK and Kwong S. Cancer incidence in the
United Farmworkers of America (UFW), 1987-1997. Am J Ind Med. 2001. 40(5): p.
596-603.
2. Mills, PK and Yang RC. Prosatte Cancer Risk in
California Farm Workers. J. Occupational and Environmental Medicine. 45(3),
249-58, 2003.
Public Health Institute
Population-Based
Epidemiology Targeted RFA, $926,325.00 / 36 mos.
(Ovarian Cancer)
During
this reporting period the data collection phase of this population-based case
control study has been completed. A total of 737 women with recently diagnosed
ovarian cancer have been identified via a rapid case ascertainment system. Of
these 595 have been met the inclusion criteria for the study and have been sent
to the interviewing subcontractor. To date 263 women, have been informed of the
research risks and have been interviewed regarding lifestyle and occupational
histories. In addition a total of 1454 age matched but cancer free women have
also been interviewed. Among cases and controls approximately 73% were
non-Hispanic white, 17% were Hispanic and 10% were of Other race/ethnicity.
After
extensive quality checks of the data were completed, preliminary analysis of
the case control data has commenced. Established risk factors for ovarian
cancer such as menstrual and reproductive history, family history of breast
and/or ovarian cancer and use of oral contraceptives have been evaluated and
have been shown to exhibit the expected relationships to ovarian cancer in this
dataset. For example we have observed an age adjusted odds ratio of 2.5 for
women who have never been pregnant indicating a two and half fold increase in
ovary cancer in women with such a history. For those women with a family
history of breast and ovarian cancer an age-adjusted odds ratio of 2.3 has been
observed. Ever use of oral contraceptives was associated with a strong
protective relationship with ovarian cancer (odds ratio=0.56). These results
suggest our methodology is sound and is capable of detecting altered risks for
ovarian cancer.
A
Job Exposure Matrix (JEM) incorporating job histories and California Department
of Pesticide Regulation (DPR) pesticide use reports has been developed and
applied to the case-control data. Very preliminary results from this JEM
approach indicate a non-statistically significant odds ratio of 1.32 for
ovarian cancer associated with exposures to the triazine herbicides.
Urinary
biomarkers of atrazine herbicides also have been analyzed for a sample of the control
women using an enzyme linked immunoassay method. Atrazine metabolites were
detected in the urine of 91% of the 45 disease free women who participated in
the biomarker substudy. However, levels of metabolites were quite low (mean
level of atrazine mercapturate =45.9 pico grams per ml) and analyses
correlating these levels with occupational and residential histories are
underway.
1. Mills PK, Riordan DG, Cress RD. Epithelial ovarian
cancer risk by invasiveness and cell type in the Central Valley of California.
Gynecol Oncol. 2004 Oct;95(1):215-25.
Allan H. Smith, M.D., Ph.D. University of California, Berkeley
Investigator-Initiated
Award, $638,803.00 / 36 mos.
(Bladder Cancer)
Arsenic is a metallic element
that occurs naturally in rocks and soils throughout the earth's crust. Under
certain conditions, arsenic can leach out of rocks and soils and contaminate
nearby rivers or ground water. Exposure to high levels of arsenic can cause
cancer of the skin, bladder, and lung.
Previous research has shown
that some people may be more susceptible to the health effects of arsenic than
others. This may be related to the way certain people process this
carcinogen. Most of the arsenic
that a person ingests is processed into a less harmful form. This processing is
not complete however, and some arsenic remains in its harmful form and may be
responsible for most of the cancer causing affects associated with arsenic. One
of the goals of this study was to investigate some of the factors that regulate
the process by which arsenic is converted to a less toxic form. This may help
us determine why some people may be more susceptible to arsenic than others. To
accomplish this goal, we measured arsenic in the urines of people exposed to
arsenic in their drinking water. These urine measurements enabled us to assess
how well each study subject processes arsenic. We compared this information to
data on subject's diet, smoking history, and other factors to see if these
factors may have played a role in arsenic processing. Another goal of this
study was to determine whether a persons ability to process arsenic may change
over time. To assess this, we collected three urine samples from each study
subject over a six-month period.
All of the data collection
portions of our study have been completed. We have collected urine and water
samples from 105 people. Over 400 samples have been collected as part of this
study. All water samples have been anlyzed for arsenic so we can see how much
arsenic people have in their water, and how much of this appears in their
urine. All urine samples have been analyzed for arsenic and its major
metabolites.
We have completed many of the
statistical tests we had planned in order to try to determine some of the factors
that control how the human body processes arsenic. To date, the factors we have
looked at include age, gender, smoking, presence of cancer or some other
long-term illness, ethnicity, and certain foods. We have also compared
methylation patterns in individuals over time, to see if methylation may change
over time within a particular person. We are finalizing analyses of several
individual components in diet such as several vitaimins, folate, protein
levels, and fruit and vegetable intake. All results are currently being
prepared for submission for publication in peer-reviewed journals.
1. Smith, AH., Lopipero, PA., Bates, MN., Steinmaus,
CS. Arsenic Epidemiology and Drinking Water Standards. Science. Vol. 296,
2145-2146, 2002
2. Smith, AH., Steinmaus, CM. Arsenic in Urine and
Water (Letter). Environmental Health Perspectives. Vol. 108 (11); A494-495,
2000
Steroid Receptor Genes and
EwingÕs Sarcoma of Bone in Children
Jan M. Van Tornout, M.D., M.S.
ChildrenÕs Hospital Los
Angeles
Pilot and Feasibility
Study, $199,321.00 / 24 mos.
(Childhood Cancer)
EwingÕs sarcoma is a cancer
that occurs mainly in children and adolescents. In the U.S. this cancer has an incidence of approximately
2.8 per million in children under the age of fifteen of all ethnicities combined
and is the second most common childhood bone tumor. Worldwide there is a very
well known difference in incidence of this disease geographically and
ethnically and some populations have a virtual absence of this cancer. The lack of this cancer in certain
populations, even after they migrate to an area that has this cancer, has been
postulated by past investigators to point to involvement of a genetic factor
rather than an environmental factor.
Although this observation has been known for some time, and much is
known about the biology of this disease, the identification of the responsible
genetic factor(s) is currently not fully explained. A second observation of the incidence of EwingÕs sarcoma
points to involvement of hormonal factors. EwingÕs sarcoma is more common in
females than males until the start of the second decade when more males develop
EwingÕs sarcoma. This pattern is
similar to the ages at which adolescents go through their pubertal growth spurt
i.e. girls begin to grow before boys, but then a few years later, boys go
through their growth spurt. This
pattern of incidence has been postulated to be related to the adolescent growth
spurt i.e. something about growth and the hormones involved could be related to
developing EwingÕs sarcoma. We
will investigate genetic differences, called polymorphisms, in those genes that
are known to be involved in pubertal growth, development and metabolism to
determine if any of these polymorphisms are associated with the disease. Different ethnicities have different
frequencies for each polymorphism.
During the period of funding
under the CCP Cycle we enrolled (1) 80 living patients with their parents
(N=135 parents) and sibings (N=100 individuals); and (2) 17 deceased patients, their parents (N=27) and
siblings (N=19 individuals) for a total of 97 patients with EwingÕs sarcoma,
162 parents and 119 siblings. The
participants were identified via the Cancer Surveillance Program of Los Angeles
County, Childrens Hospital of Los Angeles and the California Cancer
Registry. We investigated to
see whether polymorphisms in
several genes (VDR, AR, and ER, all involved in bone growth and metabolism)
were associated with EwingÕs sarcoma.
To do that we looked at the difference in the distribution of these
polymorphic genes in cases and compare the distribution from living sibling
controls and the distribution obtained by constructing hypothetical siblings
based on the parental genotypes.
Since siblings have the same ethnic background, we automatically
controlled for ethnic background.
The analysis shows that (1)
the genotype of the reeceptor substrate called IRS1, and a polymoprhic marker close to the EWS
gene may be associated with the disease itself. We are curently conducting additional analsis to validate these
findings. If validated, this study
would be the first to show a correlation between certain genetic markers and a
genetic susceptibility to develop EWS.
Alice S. Whittemore, Ph.D., M.A.
Stanford University
Investigator-Initiated
Award, $279,459.00 / 24 mos.
(Prostate Cancer)
Why do some men
with prostate cancer live healthy and productive lives well into old age, while
others develop metastases and die from their disease? Should a man with prostate cancer alter his diet, increase
his physical activity patterns, or lose weight to increase his chances of
surviving his disease?
Unfortunately, we do not know the answers to these questions, and this
continuation project was aimed at increasing our knowledge. In the period 1987-1991, we interviewed
1655 men in California, Hawaii and Canada who had a recent diagnosis of
prostate cancer. We selected these
men to represent four different ethnic groups: African-Americans, whites, Chinese-Americans and
Japanese-Americans. We asked them
about their diet and lifestyle at the time of diagnosis and ascertained their
vital status as of December 31st, 1998. We removed all identifiers and saved the information they
gave us on a computer, and now we have related this information to the menÕs
subsequent survival experience.
Our first
objective was to study certain attributes, such as race, socioeconomic status
and pre-existing medical conditions, that correlate with having a cancer that
when diagnosed has already spread beyond the prostate, or with having a cancer
that leads to death. We found that
African-Americans and foreign-born Asian-Americans were more likely than whites
to have a diagnosis of advanced cancer.
These findings held when we considered comorbidity and socioeconomic
status. The findings suggest that
efforts to improve access to and regular use of the health care system by
African-Americans and first-generation Asian-Americans may lead to earlier
prostate cancer diagnosis, when the disease is more amenable to treatment. We also observed that, among men whose
cancers were confined to their prostates, African-Americans were more likely to
die from their disease than were whites.
In contrast, survival among Asian-Americans diagnosed with localized
disease was similar to that of whites.
We also
studied how body size, physical activity patterns and dietary intakes relate to
survival from prostate cancer. One
of our major objectives was to study whether intakes of phytoestrogens and certain
fatty acids differ among men who die from their prostate cancers and men who do
not. Our findings suggest that men
with high intakes of fat, monosaturated fat, saturated fat and carbohydrates
are more likely to die of prostate cancer compared with men with low intakes of
these macronutrients. However,
phytoestrogens, fatty acids, body size and physical activity do not appear to
alter the risk of death.
1. Oakley-Girvan I, Kolonel LN, Gallagher RP, Wu AH,
Felberg A, Whittemore AS. Stage at diagnosis and survival in a multiethnic
cohort of prostate cancer patients. Am J Public Health. 2003 Oct;93(10):1753-9.
2. Oakley-Girvan I, Feldman D, Eccleshall TR,
Gallagher RP, Wu AH, Kolonel LN, Halpern J, Balise RR, West DW, Paffenbarger RS
Jr, Whittemore AS. Risk of early-onset prostate cancer in relation to germ line
polymorphisms of the vitamin D receptor.Cancer Epidemiol Biomarkers Prev. 2004
Aug;13(8):1325-30.
California Cancer Research Program
(916) 449-5550