University of Southern California Dissertations and Theses

Predicting ototoxicity evaluated by SIOP in children receiving cisplatin

(USC Thesis Other)

Predicting ototoxicity evaluated by SIOP in children receiving cisplatin

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1

Predicting Ototoxicity evaluated by SIOP in Children Receiving Cisplatin

By Ruixin Li

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN
CALIFORNIA
In Partial Fulfillment of the Requirements for the Degree
MASTER OF SCIENCE
APPLIED BIOSTATISTICS AND EPIDEMIOLOGY

May 2019

2

Acknowledgement:

I would like to express my deepest appreciation to my thesis advisor, Dr. Ha Dang, for the guidance
of my thesis and her patience. I also want to thank my thesis committee members, Dr. Meredith
Franklin and Dr. Todd Alonzo, for their amending and useful suggestions.

3

TABLE OF CONTENTS
ACKNOWLEDGEMENT ................................................................................................................................................. 2
ABSTRACT ......................................................................................................................................................................... 4
INTRODUCTION .............................................................................................................................................................. 5
DATA AND METHODS ................................................................................................................................................. 7
RESULTS .......................................................................................................................................................................... 10
DISCUSSION .................................................................................................................................................................. 18
APPENDIX ....................................................................................................................................................................... 20
REFERENCES .................................................................................................................................................................. 23

4

Abstract:

Ototoxicity induced by cisplatin therapy is a severe adverse effect, especially for children. The
primary objective of this thesis is to identify risk factors associated with severe ototoxicity
evaluated on the International Society of Pediatric Oncology Ototoxicity Scale (SIOP) in 284
eligible participants from the Children’s Oncology Group ACCL05C1 trial who received cisplatin
treatment. Severe ototoxicity evaluated by SIOP by the end of treatment was the primary outcome.
Cumulative incidence analysis was performed where deaths prior to the occurrence of primary
outcome were considered as competing events. The estimated cumulative incidence of developing
severe ototoxicity by the end of treatment was 67.35% (95% confidence interval (CI): 33.08%,
86.83%). Extended Cox models were used to assess potential risk factors with the time-dependent
covariates cisplatin attributes, receipt of nephrotoxins, and receipt of hematopoietic stem cell
transplantation (HSCT). Stepwise model selection was conducted to identify predictive risk factors
with p-value less than 0.05 in a final model. The results showed that receipt of HSCT was
statistically significantly associated with severe ototoxicity (hazard ratio (HR) = 5.97, 95%CI: 2.64,
13.51; p<0.0001). The estimated event-free probabilities at reporting period 8 were 24.74%
(95%CI: 8.87

%, 69.01%) for patients who received HSCT and 79.14% (95%CI: 71.61%, 87.45%)
for patients who never received HSCT.

5

Introduction:

Cisplatin is an essential chemotherapeutic agent for several types of cancer; however, cisplatin has
severe adverse effects, including ototoxicity. Ototoxicity is the damage to the inner ear due to
chemicals, leading to hearing loss. It can cause life-long damage to the patients’ health and life
quality, especially children. Greater than 60% of children with cisplatin therapy have developed
irreversible ototoxicity from cisplatin chemotherapy (Punnett et al., 2004).

Scientists and clinicians have been working on research to minimize and prevent the development
of ototoxicity from cisplatin therapy. One study demonstrated that trans-tympanic injections of N-
acetylcysteine could feasibly and effectively prevent cisplatin-induced ototoxicity (Riga et al.,
2013). The efficacy of amifostine for ototoxicity prevention has also been shown in a review of
several studies (Duval, 2012). Non-invasive cool water ear canal irrigation also seems to be
another highly effective otoprotective method in clinical application (Spankovich, 2016). In
addition to these promising therapies, an investigation of the risk factors related to cisplatin-
induced ototoxicity can provide insight and potential alternate strategies for ototoxicity prevention
in future trials.

The Children’s Oncology Group (COG) ACCL05C1 (ClinicalTrials.gov Identifier: NCT00458887)
is a multi-institutional observational study, aimed to investigate ototoxicity in children receiving
6

cisplatin. Based on the data of ACCL05C1, our focus is to identify risk factors associated with
cisplatin-induced ototoxicity as defined by SIOP in children by the end of treatment.

A previous study demonstrated that SIOP was the optimal criteria for evaluating ototoxicity
(Knight et al., 2017). By analyzing ACCL05C1 data, we described participants’ characteristics,
generated cumulative incidence curve and extended Cox regressions on potential risk factors.
Models that identify risk factors for ototoxicity as defined by other criteria, such as American
Speech- Language-Hearing Association Ototoxicity Criteria (ASHA), Common Terminology
Criteria for Adverse Events (CTCAE) version 3.0, the Brock Criteria (Brock), are beyond the
scope of this thesis.

7

Data and Methods:

Data:
There were 284 eligible participants enrolled in ACCL05C1 with a treatment plan containing any
cisplatin. This study was approved by the National Cancer Institute Central Institutional Review
Board (IRB) and by IRBs of participating institutions. Informed consent or assent was obtained
from participants and their guardians as appropriate.

Demographic characteristics include age at enrollment, gender, method of payment, cancer
diagnosis, and cranial irradiation at baseline. Age at enrollment was originally categorized as 0-
0.99, 1-4.99, 5-9.99, 10-14.99 and 15+ years. However, there were only 3 patients under 1 year
old with the outcome; therefore, age at enrollment was re-categorized to 0-4.99, 5-14.99 and 15+
years for all analyses. These age cut points corresponded to the 38
th
and 62
nd
percentile of the age
distribution in this study cohort. Payment methods were grouped into two categories: Medicaid,
Medicare or no means vs. others. Diagnosis was categorized based on whether a patient had any
type of brain tumor or not.

Clinical features of participants that were collected at each reporting period (i.e. follow up period)
include dose of cisplatin, duration of infusion, days of cisplatin, receipt of nephrotoxins
(aminoglycoside, loop diuretic or acetaminophen: yes or no), receipt of HSCT (yes or no),
8

relapse/progressive disease and death status. Obese (obese or not) was calculated and categorized
based on WHO Growth Chart classification for children using baseline height, weight and age at
enrollment. The first reporting period began with the begin date of the first course of chemotherapy
containing cisplatin. There were up to 13 reporting periods, which were comprised of cisplatin
courses 1-10, 4 weeks after last cisplatin dose, pre-transplant, and 4 weeks post-transplant if
patients had a transplant.

The study enrolled 122 patients, at which point study team decided to amend the case record form
to collect reporting period start and end dates. For these 122 patients, the date of initial enrollment
was regarded as the begin date of first cisplatin course. Each of their reporting periods was assumed
to last 30 days. There was no missing reporting period. One hundred and sixty-two patients were
enrolled after the forms were amended, thus they have available reporting period dates.

The outcome measure is centrally adjudicated ototoxicity by SIOP, which is graded based on
hearing thresholds >20 decibels in hearing level (dB HL) using an ordinal scale of 1 to 4, where 4
is the most severe (Brock et al., 2012). Ototoxicity evaluations were conducted at baseline, prior
to the first course of cisplatin, and repeated within one week before each subsequent cisplatin
course. The last ototoxicity evaluation was completed approximately 4 weeks after the final
cisplatin treatment, or 4 weeks after hematopoietic cell transplantation where applicable.

9

Methods:
The primary outcome of interest in this thesis project is severe ototoxicity (Grade 3-4) evaluated
by SIOP by the end of treatment, corresponding to the last available reporting period. For all
analyses, time to event was from the date of initial enrollment to the date of severe ototoxicity by
the end of treatment or death. Deaths prior to the occurrence of the primary outcome were
considered as competing events. Relapse was not considered a competing event of the primary
outcome because patients were available for ototoxicity evaluation after having a relapse. Patients
were censored by the date of the last ototoxicity evaluation or last reporting period end date,
whichever occurred later, if they did not have a primary outcome or death status.

Cumulative incidence of the primary outcome was described via competing risk analysis for the
overall study cohort. The estimated cumulative incidence by the end of reporting period 8 (243
days) and the end of therapy and corresponding 95% CIs were reported. Patients who were alive
but had missing primary outcome were considered inevaluable for these analyses.

The extended Cox model was used to assess potential risk factors where cisplatin attributes, receipt
of nephrotoxins and HSCT were evaluated as time-dependent covariates. For the time-dependent
Cox regression analysis, data were organized in counting process format. Details about the
extended Cox model is provided in the Appendix.

10

Univariate and multivariate models and the corresponding HRs with 95% CIs were obtained.
Potential risk factors for ototoxicity included age at enrollment, gender, method of payment,
diagnosis brain tumor, cranial irradiation at baseline, obese, cumulative dose of cisplatin,
cumulative duration of infusion, cumulative days of cisplatin, receipt of nephrotoxins and HSCT.
Stepwise model selection with false positive rate α=0.05 was conducted to identify the risk factors
associated with severe ototoxicity.

All tests were two-sided and p-value < 0.05 was considered statistically significant. Data were
organized and analyzed using SAS 9.4.

Results:

Demographic characteristics of this study cohort are presented in Table 1. Among 284 eligible
participants, there were 59.2% male and 29.7% participants used Medicaid, Medicare or no means
as their payment methods. Sixty-nine participants (24.3%) were diagnosed with brain tumors
(Atypical Teratoid Rhabdoid Tumor (ATRT), Medulloblastoma and Supratentorial primitive
neuroectodermal tumor (PNET)) and 15% of the participants received cranial irradiation at
baseline. Twenty seven of 284 participants (9.5%) received HSCT during ACCL05C1 study.
Forty-six participants (16.3%) were regarded as obese or extremely obese according to WHO
classification.
11

The majority of patients had 8 reporting periods (95.78%). The cumulative dose of cisplatin,
cumulative days of cisplatin infusion and cumulative days of administer were calculated over all
reporting periods for each patient. The median (range) of the cumulative dose of cisplatin based
on Body Surface Area (BSA) was 395.7 (48.3–1210.7) in mg/m2; the median (range) of the
cumulative duration of cisplatin infusion was 24 (1–480) days; and the median (range) of the
cumulative days cisplatin administered was 8 (1–30).

Table 1.Participant Characteristics of the Study Cohort (N=284)

Characteristic Mean or
Frequency
Range or
Percentage
Age at Enrollment in Years 9.7 0.1 – 21.3
0 to 0.99 6 2.1%
1 to 4.99 85 29.9%
5 to 9.99 47 16.6%
10 to 14.99 75 26.4%
15 and older 71 25.0%
Sex
Male 168 59.2%
Female 116 40.8%
Method of payment
Medicaid, Medicare or no means 84 29.7%
Other 200 70.3%
Diagnosis
Brain tumor 69 24.3%
Other 215 75.7%
Cranial radiation at baseline
12

Cumulative Incidence analysis:

Cumulative incidence of severe ototoxicity is shown in Figure 1. There were 220 patients evaluable
for the cumulative incidence analysis. Forty-seven patients (21.36%) developed severe ototoxicity
classified by SIOP grades 3 or 4, of which 34 occurred before reporting period 8. Of the 220
evaluable patients, seven deaths were observed during follow-up and considered as competing
events. The estimated cumulative incidence was 23.57% (95%CI: 16.46%, 31.43%) at reporting
period 8, which corresponded to 243 days since enrollment. By the end of therapy, the estimated
cumulative incidence was 67.35% (95%CI: 33.08%, 86.83%).

Figure 1.Cumulative incidence curve of severe ototoxicity by SIOP
Yes 42 14.8%
No 241 84.9%
Missing 1 0.3%
Obese
Yes 46 16.2%
No 238 83.8%
Median cumulative dose of cisplatin in mg/m
2
395.7 48.3 – 1210.7
Median cumulative duration of cisplatin infusion 24 1 – 480
Median cumulative days cisplatin administered 8 1 – 30
Hematopoietic stem cell transplantation
Yes 27 9.5%
No 255 89.8%
Missing 2 0.7%
13

Extended Cox Regression Analysis:
Eight hundred and seventy-five evaluable observations (i.e. reporting periods) of 215 participants
with SIOP outcome were used in the extended Cox regression analysis.

Univariate Analysis
In univariate analyses, age group at enrollment 0-4.99 years and receipt of HSCT were statistically
significantly associated with severe ototoxicity evaluated by SIOP (Table 2). Compared to the
participants enrolled at age 15+ years, the HR of developing severe ototoxicity among participants
enrolled at age 0 to 4.99 years was 2.50 (95%CI: 1.085, 5.752; p=0.0314). Receipt of HSCT had
HR of 5.97 (95%: 2.637, 13.509; p<0.0001). Other variables were not statistically significantly
14

associated with severe ototoxicity.

Table 2.Univariate extended Cox regression for potential risk factors

Potential Risk factors p-value Hazard Ratio 95% CI
Age at Enrollment in Years
0 to 4.99 0.0314* 2.499 (1.085, 5.752)
5 to 14.99 0.9995 1.000 (0.441, 2.266)
15 and older ref - -
Sex
Male 0.2099 1.495 (0.797, 2.804)
Female ref - -
Method of payment
Medicaid, Medicare or no means 0.2496 0.659 (0.324, 1.340)
Other ref - -
Diagnosis
Brain tumor 0.6725 1.140 (0.622, 2.089)
Other ref - -
Cranial radiation at baseline
Yes 0.3973 1.316 (0.697, 2.484)
No ref - -
Obese
Yes 0.8126 1.098 (0.508, 2.372)
No ref - -
Receipt of nephrotoxins
Yes 0.2728 1.379 (0.776, 2.451)
No ref - -
Cumulative dose of cisplatin in mg/m
2
0.9005 1.000 (0.998, 1.002)
Cumulative duration of cisplatin infusion 0.5497 1.001 (0.997, 1.006)
Cumulative days cisplatin administered 0.9463 1.002 (0.936, 1.074)
Hematopoietic stem cell transplantation (HSCT)
Yes <.0001* 5.969 (2.637, 13.509)
No ref - -
15

* risk factor that has p<0.05.

Multivariate Analysis
Summary statistics of a multivariate extended Cox regression model for all potential risk factors
are show in Table 3. The results indicated that age at enrollment in 0-4.99 years [2.976 (95%CI:
1.078, 8.213; p = 0.0352)], cumulative dose of cisplatin [1.003 (95%CI: 1.000, 1.005; p= 0.0259)]
and receipt of HSCT [4.441 (95%CI: 1.615, 12.207; p= 0.0039)] were significant predictors of
severe ototoxicity evaluated by SIOP in the presence of other risk factors.

Table 3.Multivariate extended Cox regression for potential risk factors

Potential risk factors Ototoxicity evaluated by SIOP
DF Parameter
Estimates
SE Chi-Sq p-value HR 95%CI
Age at Enrollment in Years
0 to 4.99 1 1.09057 0.51795 4.4333 0.0352* 2.976 (1.078, 8.213)
5 to 14.99 1 -0.11902 0.46827 0.0646 0.7994 0.888 (0.355, 2.223)
15 and older ref - - - - - -
Sex
Male 1 0.56789 0.36277 2.4506 0.1175 1.765 (0.867, 3.593)
Female ref - - - - - -
Method of payment
Medicaid, Medicare or no means 1 -0.69693 0.44975 2.4013 0.1212 0.498 (0. 206, 1.203)
Other ref - - - - - -
Diagnosis
Brain tumor 1 -0.32017 0.70551 0.2059 0.6500 0.726 (0.182, 2.894)
16

* risk factor that has p<0.05.

A stepwise model selection with false positive rate α=0.05 was conducted to identify risk factors
associated with severe ototoxicity evaluated by SIOP. This analysis showed that receipt of HSCT
was the only risk factor that remained in the final extended Cox regression model. The risk of
developing severe ototoxicity among participants who received HSCT was 5.97 times that of
participants never received HSCT (HR= 5.969, 95%CI: 2.637, 13.509; p<0.0001). The results of
the final model were given in Table 4. Age at enrollment and cumulative dose of cisplatin were
significant in the full model, however, they were eliminated from the model by stepwise selection.
One possible explanation is that regression coefficients generally depend on the presence of other
Other diagnosis ref - - - - - -
Cranial radiation at baseline
Yes 1 1.22560 0.74223 2.7266 0.0987 3.406 (0.795, 14.590)
No ref - - - - - -
Obese
Yes 1 0.46515 0.50897 0.8352 0.3608 1.592 (0.587, 4.318)
No ref - - - - - -
Receipt of nephrotoxins
Yes 1 0.20390 0.32407 0.3959 0.5292 1.226 (0.650, 2.314)
No ref - - - - - -
Cumulative dose of cisplatin in mg/m
2
1 0.00264 0.00119 4.9637 0.0259* 1.003 (1.000, 1.005)
Cumulative duration of cisplatin infusion 1 0.0007761 0.00246 0.0994 0.7526 1.001 (0.996, 1.006)
Cumulative days cisplatin administered 1 0.00912 0.04104 0.0494 0.8241 1.009 (0.931, 1.094)
Hematopoietic stem cell transplantation ( HSCT)
Yes 1 1.49080 0.51593 8.3495 0.0039* 4.441 (1.615, 12.207)
No ref - - - - - -
17

variables in a model, thus, their values are changed if one of the other variables is omitted from a
model. When one potential confounder is removed from the model, other variables’ coefficients
may move towards zero, changing from “significant” to “nonsignificant”, leading to the
elimination of other variables in the later selection steps (Heinze & Dunkler, 2017).

Table 4.Final extended Cox regression model

The estimated event-free probabilities at reporting period 8 and end of treatment stratified by
receipt of HSCT were listed in Table 5.

Table 5.Estimated event-free probabilities of severe ototoxicity by SIOP stratified by receipt of
HSCT
Estimated event-free
Probabilities
Group with HSCT (95%CI) Group without HSCT (95%CI)
At reporting period 8 0.2474 (0.0887, 0.6901) 0.7914 (0.7161, 0.8745)
End of treatment 0.0007 (4.59x10
-7
, 1.0000) 0.2970 (0.1347, 0.6548)

Risk factor Ototoxicity evaluated by SIOP
DF Parameter
Estimates
SE Chi-Sq p-value HR 95%CI
Receipt of HSCT (Yes) 1 1.78656 0.41674 18.3780 <.0001 5.969 (2.637, 13.509)
Receipt of HSCT (No) ref - - - - - -
18

Discussion:

In this thesis, ACCL05C1 data were used to investigate the cumulative incidence and risk factors
of severe ototoxicity evaluated by SIOP for children treated with cisplatin therapy.

Eleven potential risk factors were considered in the univariate and multivariate extended Cox
models, only receipt of HSCT remained in the final model. Extended Cox regression indicated that
there was prognostic effect of HSCT on severe ototoxicity grades 3 or 4 by SIOP at the end of
treatment.

Receipt of HSCT was identified as having a strong association with SIOP might because HSCT
was significantly related to neurologic complications. About 14–25% of all pediatric HSCT
patients have been reported severe neurologic complications in the context of HSCT (Brown,
2018). Ototoxicity associated with HSCT has been reported in 2.6% of all survivors of HSCT in
children (Gurney et al., 2006). Some studies have already found valid methods to reducing the
apoptosis of auditory hair cells caused by cisplatin treatment, such as a substitute of prescription
of carboplatin and 5-fluorouracil or cetuximab (Espenel et al., 2017). Efforts still need to be made
on controlling the ototoxicity owing to HSCT while the medical science is advancing. Future study
would be focusing on how to modify this risk factor to effectively minimize or prevent cisplatin-
induced ototoxicity.

19

This thesis did not consider death as a competing event for the extended Cox regression analyses.
However, results are expected to be similar given the total number of deaths was small. One
limitation of this thesis is that validation methods such as bootstrapping method of the predicted
model was not performed. Another limitation is that the predicted models for the other three types
of ototoxicity systems (ASHA, Brock, and CTCAE) were not investigated. In addition, original
reporting period dates were missing for 122 out of 284 the patients, which required us to make
assumptions and imputation regarding the duration of follow up intervals. Although they were not
addressed in this thesis, it would be of interest for future projects to verify and compare the findings
in this thesis.

20

Appendix:

Theory of Extended Cox Regression Model

Different from Cox proportional hazard regression model, the extended Cox regression model is a
modification of Cox regression model in which the proportional assumptions are not met and
contains the time-varying or interaction term with the time function.

β ：An unknown parameter. Coefficient of β represents the difference in natural log of hazard
for subject in Z2 compared to subject in Z1 at baseline or time = 0 when all the other variables are
held constant.
δ : An unknown parameter. Coefficient for time varying covariate (δ) is the change in the natural
log of the hazard ratio for Z that will occur for every unit change in time when other variables in
model are held constant.
𝜆 : An unknown function giving the hazard function for the standard set of conditions Z=0 (Cox,
1972).
p: the number of covariates.
𝑍 : Time-independent variables.
𝑍 (𝑡 ) : Time-dependent variables. The covariate path for i-th individual and it may vary over time
when we measure.
21

For individual i, suppose we have their observation time, failure indicator and other
information. The important feature of this model is that the hazard at time t depends only on the
value of the covariates at that time, Z(t). The regression effect of Z is constant β over time. The
hazard ratio exp {𝛿
Z(t)} varies over time.
Selection of time functions is according to the state of the variables used and according to the
information level of the researchers (Anavatan & Karaoz, 2013). Such as g (t) = 0, g (t) = t, g (t) =
log (t) or step function can be used in the extended Cox regression model (Husain, 2018). The
maximum likelihood estimation approach is used to estimate the parameters of the model.

Survivor function
The survivor function depends on the values of the time dependent variables over the interval from
0 to t. This is the classic formulation of the time varying Cox regression survival model
(UCSDmath).

Types of time-varying covariates
Internal covariates are related directly to individuals and can only be measured when subjects are
alive (NKY .), such as blood samples, biomarkers. External covariates contain the variables that
changes in a known way, e.g. age, time itself, and the variables that completely not related to
individuals, e.g. air temperature.
22

Inference

We still use partial likelihood to estimate β. The inference is similar to the inference in the Cox
proportion hazard regression model, except that the value of Z(t) changes at each risk set
(UCSDmath).

Procedures for model fitting
Once the time-varying variables are identified, we include them all in the extended Cox regression
model; consider whether there is main exposure or not; check functional form of covariates; assess
for interaction and confounding; do final model check (check for influential observations).
Applications of the extended Cox regression model
1. When important covariates change during a study
2. For cross-over studies, to indicate change in treatment
3. For testing the PH assumption
4. For Surrogate Outcome Analysis: An extended Cox model for time to death, with tumor
response as a time dependent covariate.
Some cautionary notes: Time-varying covariates must be carefully constructed to ensure
interpretability. There is no point adding a time-varying covariate whose value remains the same
as study time progress (NKY .).
23

References:

Anavatan, A., & Karaoz, M. (2013). Cox Regression Models with Time-Varying Covariates
Applied to Survival Success of Young Firms 1(*). Journal of Economic and Social Studies,
3(2), 53-68. doi:10.14706/JECOSS11324
Brock, P. R., Knight, K. R., Freyer, D. R., Campbell, K. C., Steyger, P. S., Blakley, B. W., . . .
Neuwelt, E. A. (2012). Platinum-induced ototoxicity in children: a consensus review on
mechanisms, predisposition, and protection, including a new International Society of
Pediatric Oncology Boston ototoxicity scale. J Clin Oncol, 30(19), 2408-2417.
doi:10.1200/JCO.2011.39.1110
Brown, V . (2018). Hematopoietic Stem Cell Transplantation for the Pediatric Hematologist/
Oncologist (V . I. Brown Ed. V ol. 24). Cham, Switzerland: Springer International
Publishing AG.
Cox, D. R. (1972). Regression Models and Life‐Tables. Journal of the Royal Statistical Society:
Series B (Methodological), 34(2), 187-202. doi:10.1111/j.2517-6161.1972.tb00899.x
Duval, M., Daniel, Sam J, and Duval, Melanie. (2012). Meta-analysis of the efficacy of amifostine
in the prevention of cisplatin ototoxicity. = Le Journal D’oto-Rhino-Laryngologie et de
Chirurgie Cervico-Faciale,. Journal of Otolaryngology -Head & Neck Surgery, 41((5)),
309–315.
Espenel, S., Garcia, M. A., Guy, J. B., Vallard, A., Ben Mrad, M., Langrand-Escure, J., . . . Magné,
24

N. (2017). [Ototoxicity in head and neck cancers after radiotherapy and chemoradiotherapy:
From primary prevention to tertiary prevention]. Cancer radiotherapie : journal de la
Societe francaise de radiotherapie oncologique, 21(1), 77-83.
doi:10.1016/j.canrad.2016.08.130
Gurney, J. G., Ness, K. K., Rosenthal, J., Forman, S. J., Bhatia, S., & Baker, K. S. (2006). Visual,
auditory, sensory, and motor impairments in long‐term survivors of hematopoietic stem
cell transplantation performed in childhood. Cancer, 106(6), 1402-1408.
doi:10.1002/cncr.21752
Heinze, G., & Dunkler, D. (2017). Five myths about variable selection. Transpl Int, 30(1), 6-10.
doi:10.1111/tri.12895
Husain, H. T., S. A.; Tahir, S.; Mukhlisin, A.; Mirna Apriani, M. (2018). The Application of
Extended Cox Proportional Hazard Method for Estimating Survival Time of Breast Cancer.
Journal of Physics: Conference Series, 979(1). doi:10.1088/1742-6596/979/1/012087
Knight, K. R., Chen, L., Freyer, D., Aplenc, R., Bancroft, M., Bliss, B., . . . Sung, L. (2017). Group-
Wide, Prospective Study of Ototoxicity Assessment in Children Receiving Cisplatin
Chemotherapy (ACCL05C1): A Report From the Children's Oncology Group. J Clin Oncol,
35(4), 440-445. doi:10.1200/JCO.2016.69.2319
NKY . (Producer). Chapter 9: Time varying (or time-dependent) covariates. .
http://ms.uky.edu/~mai/sta635/Cox%20model.pdf.
Punnett, A., Bliss, B., Dupuis, L. L., Abdolell, M., Doyle, J., & Sung, L. (2004). Ototoxicity
25

following pediatric hematopoietic stem cell transplantation: a prospective cohort study.
Pediatr Blood Cancer, 42(7), 598-603. doi:10.1002/pbc.20036
Riga, M. G., Chelis, L., Kakolyris, S., Papadopoulos, S., Stathakidou, S., Chamalidou, E., . . .
Danielides, V . (2013). Transtympanic injections of N-acetylcysteine for the prevention of
cisplatin-induced ototoxicity: a feasible method with promising efficacy. Am J Clin Oncol,
36(1), 1-6. doi:10.1097/COC.0b013e31822e006d
Spankovich, C. L., E. Ding, D. Salvi, R. Le Prell, C. G. (2016). Assessment of thermal treatment
via irrigation of external ear to reduce cisplatin-induced hearing loss. Hear Res, 332, 55-
60. doi:10.1016/j.heares.2015.11.009
UCSDmath (Producer). Lecture 7: Time-dependent Covariates in Cox Regression. .
http://www.math.ucsd.edu/~rxu/math284/slect7.pdf.

Abstract (if available)

Abstract Ototoxicity induced by cisplatin therapy is a severe adverse effect, especially for children. The primary objective of this thesis is to identify risk factors associated with severe ototoxicity evaluated on the International Society of Pediatric Oncology Ototoxicity Scale (SIOP) in 284 eligible participants from the Children’s Oncology Group ACCL05C1 trial who received cisplatin treatment. Severe ototoxicity evaluated by SIOP by the end of treatment was the primary outcome. Cumulative incidence analysis was performed where deaths prior to the occurrence of primary outcome were considered as competing events. The estimated cumulative incidence of developing severe ototoxicity by the end of treatment was 67.35% (95% confidence interval (CI): 33.08%, 86.83%). Extended Cox models were used to assess potential risk factors with the time-dependent covariates cisplatin attributes, receipt of nephrotoxins, and receipt of hematopoietic stem cell transplantation (HSCT). Stepwise model selection was conducted to identify predictive risk factors with p-value less than 0.05 in a final model. The results showed that receipt of HSCT was statistically significantly associated with severe ototoxicity (hazard ratio (HR) = 5.97, 95%CI: 2.64, 13.51

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A novel risk-based treatment strategy evaluated in pediatric head and neck non-rhabdomyosarcoma soft tissue sarcomas (NRSTS) patients: a survival analysis from the Children's Oncology Group study...

PDF

Validation of the Children’s International Mucositis Evaluation Scale (ChIMES) in pediatric cancer and SCT

PDF

Effect of biomass fuel exposure on infant respiratory health outcomes in Bangladesh

PDF

An assessment of necrosis grading in childhood osteosarcoma: the effect of initial treatment on prognostic significance

PDF

Body wall primary tumors in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Comparison for efficacy and toxicity in 2 vs 3 cycles of cisplatin consolidation therapy in children nasopharyngeal carcinoma

PDF

Surgical aortic arch intervention at the time of extended ascending aortic replacement is associated with increased mortality

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

Spatial modeling of non-tailpipe emissions and its association with children's lung function

PDF

Predicting hospital length of stay (LOS) using the National Inpatient Sample

PDF

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

PDF

An analysis of disease-free survival and overall survival in inflammatory breast cancer

PDF

Contemporary outcomes for adult congenital heart surgery in an adult tertiary care hospital

PDF

Survival of children and adolescents with low-risk non-rhabdomyosarcoma soft tissue sarcomas (NRSTS) treated with surgery only: an analysis of 234 patients from the Children’s Oncology Group stud...

PDF

Pathogenic variants in cancer predisposition genes and risk of non-breast multiple primary cancers in breast cancer patients

PDF

Instability of heart rate and rating of perceived exertion during high-intensity interval training in breast cancer patients undergoing anthracycline chemotherapy

PDF

Red and processed meat consumption and colorectal cancer risk: meta-analysis of case-control studies

PDF

Lymph node metastases in non-rhabdomyosarcoma soft tissue sarcoma

PDF

Visual acuity outcomes after cataract extraction in Chinese Americans: the Chinese American Eye Study (CHES)

PDF

The evaluation of the long-term effectiveness of zero/low fluoroscopy workflow in ablation procedures for the treatment of paroxysmal and persistent atrial fibrillation

Asset Metadata

Creator Li, Ruixin (author)

Core Title Predicting ototoxicity evaluated by SIOP in children receiving cisplatin

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Publication Date 04/17/2019

Defense Date 04/16/2019

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

Tag cisplatin,extended Cox model,HSCT,OAI-PMH Harvest,ototoxicity,SIOP

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Dang, Ha (committee chair), Alonzo, Todd (committee member), Franklin, Meredith (committee member)

Creator Email re_422@outlook.com,ruixinli@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-139122

Unique identifier UC11676907

Identifier etd-LiRuixin-7200.pdf (filename),usctheses-c89-139122 (legacy record id)

Legacy Identifier etd-LiRuixin-7200.pdf

Dmrecord 139122

Document Type Thesis

Format application/pdf (imt)

Rights Li, Ruixin

Type texts

Source 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 a...

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