Melanoma Comorbidity Clusters
Principal Investigator
Name
Vishnutheertha Kulkarni
Institution
Johns Hopkins University
Position Title
Student
Email
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCO-869
Initial CDAS Request Approval
Nov 30, 2021
Title
Melanoma Comorbidity Clusters
Summary
Melanoma incidence has been increasing in the United States over the past few years. The median age of US adults diagnosed with melanoma hovers around 60 years. The elderly have increased comorbidities and a higher prevalence of chronic diseases such as diabetes and heart disease. These conditions may play a role in cancer detection and treatment, thereby affecting prognosis of melanoma patients. Additionally, melanoma treatments may exacerbate the severity of these pre-existing comorbidities, resulting in worse prognoses. However, few studies have explored the interactions and associations between melanoma and comorbidities.
Thus, there is an increasing need to investigate the impact of comorbidities on melanoma patients including response to treatment, survival rates, risks of complications, and quality of life. We aim to identify subgroups of melanoma patients with different comorbidities based upon clustering algorithms, such as latent class analysis. Patients’ comorbidity burden will be quantified using methods such as the Charlson Comorbidity Index (CCI) as well as the individual comorbidies contributing to the score (e.g. myocardia infraction, diabetes, chronic pulmonary disease). Additionally, diseases not captured in the CCI, including obesity and autoimmune conditions, will be analyzed. Specifically, we will examine the prevalence rates of the most common comorbidities in patients with melanoma and any associations between comorbidities and patient demographics, including age, gender, melanoma stage, and treatment. We predict that comorbidity clusters can be identified using clustering algorithms, and that patients in different comorbidity clusters will have different outcome profiles.
Thus, there is an increasing need to investigate the impact of comorbidities on melanoma patients including response to treatment, survival rates, risks of complications, and quality of life. We aim to identify subgroups of melanoma patients with different comorbidities based upon clustering algorithms, such as latent class analysis. Patients’ comorbidity burden will be quantified using methods such as the Charlson Comorbidity Index (CCI) as well as the individual comorbidies contributing to the score (e.g. myocardia infraction, diabetes, chronic pulmonary disease). Additionally, diseases not captured in the CCI, including obesity and autoimmune conditions, will be analyzed. Specifically, we will examine the prevalence rates of the most common comorbidities in patients with melanoma and any associations between comorbidities and patient demographics, including age, gender, melanoma stage, and treatment. We predict that comorbidity clusters can be identified using clustering algorithms, and that patients in different comorbidity clusters will have different outcome profiles.
Aims
Identify subgroups of melanoma patients with differences in comorbidity profiles using clustering algorithms.
Study the differences in outcomes of melanoma patients with different comorbidity profiles.
Determine the accuracy of the clustering algorithm to stratify melanoma patients into risk groups based upon their comorbidity profiles.
Collaborators
Shannon Wongvibulsin, University of California Los Angeles
Jeffrey Scott, Johns Hopkins University