University of Southern California Dissertations and Theses

The interplay between tobacco exposure and polygenic risk score for growth on birthweight and childhood acute lymphoblastic leukemia

(USC Thesis Other)

The interplay between tobacco exposure and polygenic risk score for growth on birthweight and childhood acute lymphoblastic leukemia

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Copyright 2023 Vikas Kunta
The Interplay Between Tobacco Exposure and Polygenic Risk Score for
Growth on Birthweight and Childhood Acute Lymphoblastic Leukemia
By
Vikas Raj Kunta
A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(APPLIED BIOSTATISTICS AND EPIDEMIOLOGY)
December 2023

ii
"Science is not everything, but science is very beautiful.”
- J. Robert Oppenheimer

iii
To My Mom, Dad, and Sister

iv
Acknowledgements
Greatly appreciate the help of Dr. Joseph Wiemels and all the professors
I have had the pleasure of working with during my time at USC.
Thank You.

v
Table of Contents
Epigraph………………………………………….……………………………......ii
Dedication……………………………………………………………...…………iii
Acknowledgments…………………………………………………………...……iv
List of Tables……………………………………………...…………….………...vi
List of Figures………………..…………………………………………………..vii
Abstract………………………………………………………….……..………..viii
Introduction ………………………………………………………...………….....1
Chapter 1: Methods………………………………….………………....……........4
Chapter 2: Results…………………………………………………………...........6
Chapter 3: Discussion…………………………………….……………………...17
Bibliography……………………………………………………………………...20
Appendix………………………………………….....…………………………....22

vi
List of Figures:
Figure 1: Case/Control Distribution of European Ancestry Group ..…..........10
Figure 2: Case/Control Distribution of Latino Ancestry Group ………..........10
Figure 3: Birthweight Distribution of European Ancestry Group ……..…....11
Figure 4: Birthweight Distribution of Latino Ancestry Group ……..........…..11
Figure 5: PRS Distribution of European Ancestry Group ...............…….…...12
Figure 6: PRS Distribution of Latino Ancestry Group ……….........................12
Figure 7: PRS Density Plot of European Ancestry Group ………...................13
Figure 8: PRS Density Plot of Latino Ancestry Group ……….........................13
Figure 9: Fractional Abundance Distribution of European Ancestry Group.14
Figure 10: Fractional Abundance Distribution of Latino Ancestry Group.....14
Appendix Figure 1: SNP Loci Reference Data Table.........................................22
Appendix Figure 2: Race Distribution in the State of California in 2018........23

vii
List of Tables:
Table 1: Descriptive Statistics of European Ancestry Group………..…..........15
Table 2: Descriptive Statistics of Latino Ancestry Group……..….....…..........15
Table 3: Linear Regression Models with Birthweight as an Outcome …........16
Table 4: Logistic Regression Models with Case Status as an Outcome ...........16

viii
Abstract
Background
The data collected from CCRLP (California Cancer Registry Linkage Project) was used
in efforts to understand what is potentially causing cancer diagnoses in children based on
different predispositions to genetic, biological, and environmental events. Here we explore the
relationships between genetic propensity to higher birthweight (a risk factor for ALL) and
pregnancy tobacco exposure (a carcinogen that decreases birthweight).
Methods
From 1988-2011, children’s acute lymphoblastic leukemia cases were tracked in the state
of California and were later matched to controls that shared similar traits as the cases without
ever having cancer. Using that, we divided the population provided into two major ethnic groups:
European and Latino ancestry individuals. The two primary outcomes for this study were
birthweight and cancer diagnosis so a linear regression was used for birthweight and a logistic
regression was used for cancer diagnosis. An interaction effect between tobacco smoking and
birthweight PRS was assessed, as they are two risk factors with opposing effects on each other.
Results
In the European portion of the study, there were 1,022 cases and 1,069 controls and the
mean birthweight and PRS were both higher in the cases (3,464.72g, 18.9 score) than the
controls (3,426.51g, 16.85 score). In the Latino portion of the study, there were 1,280 cases and
1,400 controls measured and, similar to the European study, the mean birthweight and PRS
scores were higher in cases (3,441.07g, 18.97 score) than controls (3,383.61g, 17.18 score) with
the median birthweights showing a similar increase for the cases as well. European cases and
controls differed for PRS in the linear model for birthweight as an outcome whereas the smoking

ix
(FA) was statistically significant. Latino results for this linear regression model were similar with
the difference between smoking was not significant for the cases, just the controls.
In the logistic regression model done with case status as the outcome, it was found that
both the European and Latino populations had a statistically significant relationship of PRS with
case status as the outcome. The distinction found was that Latino population’s birthweight was
statistically significant whether the outcome was a case or control. Throughout the analysis, it
was found the multivariable seemed to not make a significant difference vs. cases whether they
are univariate or multivariable.
Conclusion
The results showed that accumulative PRS does influence case status and can potentially
increase one’s chances of contracting childhood acute lymphoblastic leukemia for both European
and Latino groups. Birthweight also appears to significantly influence case status for the Latino
groups but was found to be insignificant for European groups. Birthweight was influenced by
smoking in European cases and controls and Latino controls. The interaction of smoking and
birthweight were found to be significant univariately in both European and Latino groups with
case status as an outcome. For the multivariable including the interaction and case status as an
outcome, there were only two variables that were significant being the European PRS score and
the Latino birthweight. When birthweight was the outcome for the interaction variable, there
were no significant associations in the multivariable including all the variables.

1
Introduction
Childhood acute lymphoblastic leukemia (ALL) is a form of cancer where the bone
marrow makes too many immature lymphocytes which reduces and takes the place of healthy
white blood cells, red blood cells, and platelets and is the most common leukemia found in
children, with up to 80% of children’s leukemia cases are classified as ALL, according to the
National Cancer Institute (NCI). This can cause symptoms such as pneumonia, easy bleeding,
getting sick frequently, fatigue, and easy bruising depending on the age of the child and how far
the disease has progressed (Dana-Farber). The American Cancer Society states that “the 5-year
relative survival rate for children 0 to 14 with ALL is 92%”. Children and their parents are
typically presented with multiple treatment options with chemotherapy being the most common
method of treatment and sometimes targeted therapy or radiation therapy are used alongside
chemotherapy depending on the status of the cancer and the patient, according to the Mayo
Clinic. They typically wait until 5 years of the patient being disease free as the cancer has a
higher chance of coming back within those 5 years than once after the 5 years has completed, in
which the cancer is unlikely to come back. Studies have shown that some minority communities
(particularly Latinos) are more likely to get diagnosed with diseases and are more likely to suffer
serious effects of those diseases, hence the desire to run this study with two major ethnic groups
in the state of California including Latinos – the largest birth population in that state.
In multiple recent studies, the effects of birthweight on childhood acute lymphoblastic
leukemia have been explored and documented in showing that it is likely that a high birthweight
can increase one’s chances of developing cancer later in life. Gestational age is a way to measure
how far along the unborn child is during pregnancy and is considered to be normal when the
pregnancy is at 38-42 weeks at birth, therefore, babies born at or before 37 weeks are to be

2
considered premature. This is accounted for in the analysis since it is an influential predictor of
birthweight as the longer the unborn child is in the womb (gestation), the higher the birthweight
can be, hence the need for gestational age to be controlled when analyzing the effects of
birthweight in this study. The effects of tobacco smoking by the mother were also investigated as
an exposure in this study due to its potential effect on birthweight. Tobacco smoking during
pregnancy has been shown to decrease the birthweight of a child and causing a very low
birthweight which can also potentially affect their risk of getting ALL at some point in the future.
Therefore, birthweight (adjusted for gestational age) and tobacco exposure will be measured and
referred to as FA (fractional abundance of a DNA methylation mark at the gene AHRR) will be
assessed in the study along with PRS values.
Polygenic risk scores (PRS) are an accumulation of the effects of genetic variants in each
person’s genome that is used to measure and estimate the chance of developing a particular
disease or a specific trait. The higher the score is, the more likely one is to contract the disease
and the lower the score is, the less likely someone is to contract the disease, according to the
Center for Disease Control (CDC). Genes can be influenced by ethnicity as those in similar
ethnic groups tend to have the similar traits and genes passed down from common ancestors and
be exposed to different environmental interactions throughout generations, potentially causing
predispositions to various diseases. For this, we utilized SNPs (single nucleotide polymorphism)
to base our calculations on. The approximately 50 SNPs that were used in this analysis are all the
loci that relate to birthweight at a position in the DNA of an individual. Due to these factors, we
utilized one large dataset from the California Cancer Registry Linkage Project (CCRLP) data and
divided it into two races: Non-Hispanic European ancestry and Latino ancestry individuals for
this analysis.

3
The hypotheses being tested individually are the effects of tobacco smoking and
birthweight PRS on both case status and birthweight in addition to the interaction of tobacco
smoking with birthweight PRS on those two outcomes in both univariate and multivariable
combinations. We expect the higher the PRS, the higher the birthweight will be while the more
tobacco exposure there is, the lower the birthweight will be, and we will see what their effects
will be on case status along with the interaction of the two.

4
Chapter 1: Methods
Study Design
In this analysis, a genetic linkage-based study was done based on matched case-control
data collected from various children’s cancer hospitals in the state of California from 1991 to
2011 consisting of childhood acute lymphoblastic leukemia cases and separate controls. The
cases and the controls were used to match against a list of approximately 60 different loci of
SNPs (single nucleotide polymorphism) pertaining to birthweight, of which the about 50 were
used due to no matches in some of the SNPs in the participants being used in this study. The
polygenic risk scores were based on a study done by Horikoshi, et. al and was further calculated
by the summation of polygenic risk scores, calculated in this study, for use in this analysis.
Database
The California Cancer Registry Linkage Project (CCRLP) data was used for this analysis
and was divided it into two primary races: Non-Hispanic European ancestry and Latino ancestry
individuals. This was done as they were the two largest ethnic groups in the study, and both have
different genetic backgrounds and exposures that could have influenced their DNA and genes
throughout the generations. The data was collected for 20 years, and the cases were matched with
the controls by utilizing birth records from the California Department of Public Health. This
database was overseen by the Center for Disease Control and Prevention (CDC) and has assisted
California for many decades now in effort to assist the state in improving survivorship of cancer,
according to the California Department of Public Health. As this is a matched case-control study,
the cases where matched with controls based on biological sex, month/year of birth, and one of
the two race/ethnicity groups chosen for this analysis. The information collected was the case
status, birthweight, gestational age, and the fractional abundance in regard to tobacco smoking

5
exposure. The SNPs (single nucleotide polymorphism) were collected from a research paper
from a data table (Appendix Figure 1) provided showcasing the loci found to be associated with
birthweight in almost 300,000 European ancestry and Trans-ancestry individuals (Horikoshi).
Statistical Software and Tests
The study this paper is based on primarily utilized RStudio Version 2023.06.1+524
(RStudio Team (2020). RStudio: Integrated Development for R. RStudio, PBC, Boston, MA) for
analyses. Linear and logistic regression models were utilized, and significance was used at a 95%
confidence interval, meaning a p-value of <0.05 was the baseline in knowing whether a result
was statistically significant or not. The logistic regression model was used for when case-control
was the desired outcome as it is a dichotomous outcome, meaning only 2 possible outcomes. The
linear regression model was used for when birthweight was the desired outcome as birthweight is
deemed a continuous variable.

6
Chapter 2: Results
In this study, all figures and tables are done twice as the European and Latino ancestry
groups are kept separate in every step of the analysis. In figures 5 and 6, the distribution of PRS
scores is shown for both European and Latino groups, respectively, and the common trend seen
there is that the controls starts off with higher frequencies in the lower PRS levels and once a
higher threshold is crossed, the cases are the ones with a higher frequency in the higher PRS
levels. In the density plots showcased in figures 7 and 8, one can see that a majority of PRS
values are in the middle, showing a normal distribution which is typically how PRS values are
spread throughout a population. In figures 1 and 2, the distribution of cases and controls of
European and Latino groups, respectively, are shown in a bar graph and as seen, there are a little
more controls than cases which is what was needed for this study in effort to better control
potential confounders and better understand the effects of birthweight and cancer diagnosis in the
populations that are being studied.
Birthweight as an Outcome
In the European portion of the study, there were 1,022 cases and 1,069 controls and the
mean birthweight and PRS were both higher in the cases (3,464.72g, 18.9 score) than the
controls (3,426.51g, 16.85 score). However, the median birthweights for both cases and controls
were very similar to each other with the difference being just 2 grams. The median in the PRS,
however, remained higher in the cases than controls (18.90 vs. 16.85) and the data described is
shown in Table 1. In the Latino portion of the study, there were 1,280 cases and 1,400 controls
measured and, similar to the European study, the mean birthweight and PRS scores were higher
in cases (3,441.07g, 18.97 score) than controls (3,383.61g, 17.18 score) with the medians
showing a similar increase for the cases as well, shown in Table 2.

7
In Table 3, linear regression models were conducted with birthweight as an outcome for
our targeted variables. One column was done univariately for our desired variables, and another
was a summation of the multivariable model consisting of all the variables we want included
with birthweight continuing to be the outcome, separated for European and Latino populations.
One thing to keep in mind, birthweight in this model was always calculated with the effect of
gestational age considered, in effort to reduce confounding.
The European controls in this model showed an insignificant difference in PRS scores for
both the univariate and multivariable models against birthweight alone and with birthweight and
the other variables mixed in, as the p-values were greater than 0.05 and the effect estimates were
not much different. The European cases for PRS scores showed similar results as the controls did
with insignificant p-values and not much of a change in effect estimates as it is within the range
of the standard error. However, smoking for both cases and controls in the European univariate
and multivariable models showed a significant effect between smoking and birthweight as the pvalues were statistically significant (<0.001) regardless of other variables being involved. The
controls had a statistically significant univariate interaction variable with a p-value of 0.0136 and
effect estimate of 0.165. In the multivariable column of the table that includes interaction, there
were no statistically significant relationships for both cases and controls in the European group.
In the Latino proportion of this birthweight model, the results were similar to the
European model with a discrepancy in the smoking aspect of the analysis where the cases were
not statistically significant but the control group’s fractional abundance (smoking) was
statistically significant with birthweight in both the univariate and multivariable models,
however, the results were basically the exact same in the multivariable and univariate model for
the “Latino Controls Smoking (FA)” group. In the Latino group, the interaction of PRS and

8
smoking, did not have any statistically significant relationships presented in both univariate and
multivariable (with interaction) models.
Leukemia Case Status as an Outcome
In Table 4, logistic regression models were conducted with case status as the outcome for
our targeted variables. One column was done univariately for our desired variables, and another
was the multivariable model with case/control continuing to be the outcome, separated for
European and Latino populations.
For the European population, when the logistic regression model ran the cases and
controls with the PRS, there was a statistically significant difference with cases and controls and
the PRS presented, although the effect size remained the same regardless of the additional
variables added in the multivariable model. Smoking, gestational age, and birthweight were not
statistically significant between cases and controls in both univariate and multivariable models.
The interaction model of PRS and smoking for the European group were statistically significant
(P = <0.001) in the univariate model with a correlation coefficient of 0.00218. In the
multivariable model including the interaction effect, there was only one significant variable
which was the PRS score for Europeans as it had a correlation coefficient of 0.512 with a p-value
of 0.00897.
The Latino group showed similar results for smoking and gestational age as they both
were not statistically significant. The PRS score, however, there appears to be a statistically
significant difference for cases and controls. Unlike the European group, birthweight was
significantly different in cases and controls in the Latino population of this study, with the effect
estimates similar to one another in both univariate and multivariable settings, showing not much
of an influence with other variables in this study. The interaction model of PRS and smoking for

9
the Latino group were statistically significant (P = <0.001) in the univariate model with a
correlation coefficient of 0.0022. In the multivariable model that includes the interaction, there
was only one significant variable which was birthweight with a correlation coefficient of 0.19
and a p-value of 0.023.

10
Figure 1: Case/Control Distribution of European Ancestry Group
Figure 2: Case/Control Distribution of Latino Ancestry Group

11
Figure 3: Birthweight Distribution of European Ancestry Group
Figure 4: Birthweight Distribution of Latino Ancestry Group

12
Figure 5: PRS Distribution of European Ancestry Group
Figure 6: PRS Distribution of Latino Ancestry Group

13
Figure 7: PRS Density Plot of European Ancestry Group
Figure 8: PRS Density Plot of Latino Ancestry Group

14
Figure 9: Fractional Abundance Distribution of European Ancestry Group
Figure 10: Fractional Abundance Distribution of Latino Ancestry Group

15

16

17
Chapter 3: Discussion
This study was a result of an accumulation of 20 years of data collected by the California
Cancer Registry Linkage Project and the birth records collected from the California Department
of Public Health and overseen by the CDC. The cases were verified and kept up to date and other
factors with their matched controls. The two races that were used for this study were two of the
most populous in California with Hispanics representing about 39% and non-Hispanic Whites
representing about 37% of the state’s demographics (Appendix Figure 2). As they were the
biggest sample size that the state could provide, the two groups for this study were chosen and
analyzed.
The results showed that the European cases and controls for smoking with birthweight as
an outcome showed a p-value less than 0.001 for both categories in both univariate and
multivariable linear regressions done. In those two variates, there was not a significant change in
effect size nor p-value but did seem to suggest that birthweight increases by 8.014g for cases and
8.419g for controls for each unit increase of fractional abundance. This trend was similar in
Latino controls (10.293g, p-value = <0.001) for smoking (FA) against birthweight in the linear
regression model. However, it was not significant in Latino cases (p-value = 0.219). This would
make sense since the birthweight should go up when there is less tobacco exposure because that
means the fractional abundance (FA) increases as well. The lower the FA means that there is
more exposure to smoking. There was not a significant relationship between PRS score and
birthweight for both European and Latino groups with p-values, shown in Table 3, exceeding the
threshold of 0.05 on all accounts relating to the two variables in univariate and multivariable
forms.

18
With case status as the outcome, both European and Latino groups showed a positive
correlation and statistically significant p-values supporting the relationship between PRS and
case status with p-values less than 0.001. The European PRS group showed an increase of 0.235
and the Latino PRS group showed an increase of 0.214 in contracting a case. The Latino group
was the only one that had another significant relationship in this model and that was with
birthweight with case status as an outcome. The higher the birthweight, the higher chance of
developing a case of cancer with a coefficient of 0.000197, which is quite small and a p-value of
0.013. Once again, the multivariable model was not different and showed no effects with all the
variables accounted for together.
Limitations of this study are the data includes just 2 ethnic groups. Although they are the
two most common and highest populated groups in the state, this study can be improved by
going on a national scale and including those of African American, East Asian, South Asian,
Native American, and Middle Eastern backgrounds. For this, we would need a verified and
reliable database consisting of cases and controls and need to be calculated for a significant
period of time such as this one was. This would allow us to see if there are more discrepancies in
race and could take it to another level by not just looking at potential different genetic
predispositions that some groups may have to certain diseases, but also seeing if parent/primary
caregiver’s economic status can influence the data. Dietary habits of the mother could be
included as well but would be difficult to confirm the authenticity of those results. Another
limitation of this study was that not all the loci of SNPs were utilized (approximately 8-10 were
missing) due to none of the participants matching for it. This could be a biasing factor as the
missing loci could have had an influence in the outcome of this data. Further research might need

19
to be done to see if those missing loci are rare in these two ethnic groups and potentially more
common in another race group that would be included in a bigger study, as mentioned earlier.
The interaction terms in this study yielded potentially provocative results primarily in
table 4 when presented univariately against case status as both the European and Latino groups
were statistically significant. Only the PRS score for Europeans and the Latino birthweight in
their independent multivariable with interaction models were found to be significant. When
interaction was measured against birthweight, the multivariable models produced no significant
results but the European Controls interaction, univariately, was found to be statistically
significant, suggesting more research could be explored on this matter since the Latinos did not
have this anywhere. It should be noted that the cases for European were very close to being
significant with a p-value of 0.057, hence there could be a potential study done on Europeans’
predisposition to this interaction term of PRS with smoking.
To summarize, the multivariable models were not as effective as desired in this analysis
and the effective results were reflected by univariate results with the variables individually
measured against the outcome of birthweight and case status. PRS score of both European and
Latino groups influenced case status with higher the PRS, the more likely to be a case. For
Latinos, the higher the birthweight, it was significantly likely to be a case as well. In the
birthweight model, it was shown in the European group that birthweight increases by 8.014g for
cases and 8.419g for controls for each unit increase of fractional abundance and showed similar
results in Latino controls only. Interaction term was found to be significant univariately against
case status for both groups while having little impact in the multivariable model for both
birthweight and case status separately.

20
Bibliography
Centers for Disease Control and Prevention. (2022, May 20). Polygenic risk scores. Centers for
Disease Control and Prevention.
https://www.cdc.gov/genomics/disease/polygenic.htm#:~:text=Polygenic%20risk%20score
s%20for%20a,in%20the%20middle%20(orange).
Childhood acute lymphoblastic leukemia treatment. National Cancer Institute. (n.d.).
https://www.cancer.gov/types/leukemia/patient/child-all-treatmentpdq#:~:text=and%20treatment%20options.-
,Childhood%20acute%20lymphoblastic%20leukemia%20(ALL)%20is%20a%20type%20o
f%20cancer,the%20blood%20and%20bone%20marrow.
Childhood acute lymphoblastic leukemia. Dana. (n.d.). https://www.dana-farber.org/childhoodacute-lymphoblastic-leukemia/
Fuentes-Paez, G., Escaramís, G., Aguilar-Lacasaña, S., Andrusaityte, S., Brantsæter, A. L.,
Casas, M., Charles, M.-A., Chatzi, L., Lepeule, J., Grazuleviciene, R., Gützkow, K. B.,
Heude, B., Maitre, L., Ruiz-Arenas, C., Sunyer, J., Urquiza, J., Yang, T. C., Wright, J.,
Vrijheid, M., … Bustamante, M. (2022, April 4). Study of the combined effect of maternal
tobacco smoking and polygenic risk scores on birth weight and body mass index in
childhood. Frontiers. https://www.frontiersin.org/articles/10.3389/fgene.2022.867611/full
Health, D. of P. (n.d.). California Cancer Registry.
https://www.cdph.ca.gov/Programs/CCDPHP/DCDIC/CDSRB/Pages/California-CancerRegistry.aspx
Horikoshi, M., Beaumont, R. N., Day, F. R., Warrington, N. M., Kooijman, M. N., FernandezTajes, J., Feenstra, B., van Zuydam, N. R., Gaulton, K. J., Grarup, N., Bradfield, J. P.,
Strachan, D. P., Li-Gao, R., Ahluwalia, T. S., Kreiner, E., Rueedi, R., Lyytikäinen, L. P.,
Cousminer, D. L., Wu, Y., Thiering, E., … Freathy, R. M. (2016). Genome-wide
associations for birth weight and correlations with adult disease. Nature, 538(7624), 248–
252. https://doi.org/10.1038/nature19806
Mayo Foundation for Medical Education and Research. (2022, September 21). Acute
lymphocytic leukemia. Mayo Clinic. https://www.mayoclinic.org/diseasesconditions/acute-lymphocytic-leukemia/diagnosis-treatment/drc20369083#:~:text=Chemotherapy.,the%20consolidation%20and%20maintenance%20phas
es.
RStudio Team (2020). RStudio: Integrated Development for R. RStudio, PBC, Boston, MA URL
http://www.rstudio.com/.
Wiemels, J. L., Walsh, K. M., de Smith, A. J., Metayer, C., Gonseth, S., Hansen, H. M., Francis,
S. S., Ojha, J., Smirnov, I., Barcellos, L., Xiao, X., Morimoto, L., McKean-Cowdin, R., Wang,

21
R., Yu, H., Hoh, J., DeWan, A. T., & Ma, X. (2018). GWAS in childhood acute lymphoblastic
leukemia reveals novel genetic associations at chromosomes 17q12 and 8q24.21. Nature
communications, 9(1), 286. https://doi.org/10.1038/s41467-017-02596-9

22
Appendix
Appendix Figure 1: SNP Loci Reference Data Table (Horikoshi, et. al)

23
Appendix Figure 2: Race Distribution in the State of California in 2018 (U.S Census Bureau)

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The influence of DNA repair genes and prenatal tobacco exposure on childhood acute lymphoblastic leukemia risk: a gene-environment interaction study

PDF

Using genetic ancestry to improve between-population transferability of a prostate cancer polygenic risk score

PDF

Age related macular degeneration in Latinos: risk factors and impact on quality of life

PDF

Construction of a surgical survival prediction model of stage IV NSCLC patients-based on seer database

PDF

Disparities in colorectal cancer survival among Latinos in California

PDF

Adipokines do not account for the association between osteocalcin and insulin sensitivity in Mexican Americans

PDF

Association of traffic-related air pollution and lens opacities in the Los Angeles Latino Eye Study

PDF

Fish consumption and risk of colorectal cancer

PDF

Exploring the interplay of birth order and birth weight on leukemia risk

PDF

The risk estimates of pneumoconiosis and its relevant complications: a systematic review and meta-analysis

PDF

Disparities in exposure to traffic-related pollution sources by self-identified and ancestral Hispanic descent in participants of the USC Children’s Health Study

PDF

Native American ancestry among Hispanic Whites is associated with higher risk of childhood obesity: a longitudinal analysis of Children’s Health Study data

PDF

Genetic studies of cancer in populations of African ancestry and Latinos

PDF

Predictive factors of breast cancer survival: a population-based study

PDF

Genetic epidemiological approaches in the study of risk factors for hematologic malignancies

PDF

Risk factors and survival outcome in childhood alveolar soft part sarcoma among patients in the Children’s Oncology Group (COG) Phase 3 study ARST0332

PDF

Ancestral/Ethnic variation in the epidemiology and genetic predisposition of early-onset hematologic cancers

PDF

The impact of the COVID-19 pandemic on cancer care delivery

PDF

Risk factors of pelvic floor disorders in the multiethnic cohort study

PDF

A multivariate approach to explore nursing home characteristics contributing to high performance on CMS' publicly reported quality measures

Asset Metadata

Creator Kunta, Vikas Raj (author)

Core Title The interplay between tobacco exposure and polygenic risk score for growth on birthweight and childhood acute lymphoblastic leukemia

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Degree Conferral Date 2023-12

Publication Date 10/30/2023

Defense Date 10/27/2023

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag birthweight,cancer,leukemia,OAI-PMH Harvest,polygenic risk scores,smoking

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Wiemels, Joseph (committee chair), Choudhury, Farzana (committee member), Gauderman, William (committee member)

Creator Email vkunta@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113763044

Unique identifier UC113763044

Identifier etd-KuntaVikas-12442.pdf (filename)

Legacy Identifier etd-KuntaVikas-12442

Document Type Thesis

Format theses (aat)

Rights Kunta, Vikas Raj

Internet Media Type application/pdf

Type texts

Source 20231103-usctheses-batch-1104 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu