Kaelyn C. Cummins
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Kaelyn C. Cummins has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Mohamad El Moheb
Department of Surgery University of Virginia Charlottesville Virginia USA
11 shared publications - Russell Gardner Witt
Department of Surgery University of Virginia Charlottesville Virginia USA
11 shared publications - Allan Tsung
Department of Surgery, University of Virginia, Charlottesville, Virginia, USA
10 shared publications - Carolyn Tsung
Department of Surgery, University of Virginia, Charlottesville, Virginia
10 shared publications - Chengli Shen
Division of Surgical Sciences, Department of Surgery (R.T., Y.H., T.S., J.L., C.S., X.Z., J.M., K.C.K., L.-W.G.), School of Medicine, University of Virginia, Charlottesville
10 shared publications - David R. Brenin
Department of Surgery, University of Virginia, Charlottesville, Virginia
10 shared publications
Latest Publications
- Open Access
- 14 July 2026
- Published by Wiley in Pigment Cell & Melanoma Research
- Vol. 39 (4) , e70112
Characterizing the Mutational Landscape of In‐Transit Melanoma Metastases
In‐transit melanoma (ITM) is a unique presentation of metastatic cutaneous melanoma associated with therapeutic challenges. Despite its clinical importance, the molecular drivers of ITM remain poorly defined. We aimed to characterize the mutational landscape of ITM and identify genetic alterations distinguishing it from other melanoma metastases. Tumor samples from 528 patients in the MSK‐IMPACT dataset, comprised of over 300 cancer‐related genes, were analyzed. Samples were classified as primary, in‐transit, regional lymph node, or distant metastases. Driver mutation frequencies were compared across groups, and mutual information (MI) and principal component analysis (PCA) were used to identify ITM‐associated genes and mutation patterns. NRAS Q61 mutations were enriched in ITM compared with other melanoma sites, while NF1 mutations were less common. Exploratory MI, PCA, and pairwise analyses identified recurrent NRAS MUT /wild‐type gene patterns that may distinguish ITM from other melanoma samples. Genes retained in the wild‐type state alongside NRAS MUT were associated with PI3K/AKT/mTOR, TGF‐β, and E2F‐related pathways. Collectively, these findings suggest that ITM is enriched for NRAS Q61 mutations and may have a comparatively lower co‐mutation burden, providing a basis for future studies of ITM biology and clinical behavior.
- 14 July 2026
- Published by Wiley in Pigment Cell & Melanoma Research
- Vol. 39 (4) , e70112
Open AccessCharacterizing the Mutational Landscape of In‐Transit Melanoma Metastases
In‐transit melanoma (ITM) is a unique presentation of metastatic cutaneous melanoma associated with therapeutic challenges. Despite its clinical importance, the molecular drivers of ITM remain poorly defined. We aimed to characterize the mutational landscape of ITM and identify genetic alterations distinguishing it from other melanoma metastases. Tumor samples from 528 patients in the MSK‐IMPACT dataset, comprised of over 300 cancer‐related genes, were analyzed. Samples were classified as primary, in‐transit, regional lymph node, or distant metastases. Driver mutation frequencies were compared across groups, and mutual information (MI) and principal component analysis (PCA) were used to identify ITM‐associated genes and mutation patterns. NRAS Q61 mutations were enriched in ITM compared with other melanoma sites, while NF1 mutations were less common. Exploratory MI, PCA, and pairwise analyses identified recurrent NRAS MUT /wild‐type gene patterns that may distinguish ITM from other melanoma samples. Genes retained in the wild‐type state alongside NRAS MUT were associated with PI3K/AKT/mTOR, TGF‐β, and E2F‐related pathways. Collectively, these findings suggest that ITM is enriched for NRAS Q61 mutations and may have a comparatively lower co‐mutation burden, providing a basis for future studies of ITM biology and clinical behavior.
- Open Access
- 17 June 2026
- Published by Wiley in Cancer Medicine
- Vol. 15 (6) , e72043
Nationwide Analysis of Sociodemographic Disparities in Health‐Related Quality of Life of Colon Cancer Survivors
Background: Health‐related quality of life (HRQOL) measures patients' overall well‐being and can help guide survivorship care. Evidence comparing individual patients' HRQOL pre‐ and post‐diagnosis is limited due to the unpredictable nature of cancer onset. Our study aims to evaluate trends in HRQOL in elderly colon cancer patients and identify predictors of HRQOL change across two key transitions: (1) pre‐ to post‐diagnosis, and (2) at diagnosis versus 2 years later, with the goal of identifying at‐risk patients for a significant HRQOL decrease. Methods: Utilizing the SEER‐Medicare Health Outcomes Survey (SEER‐MHOS), patients diagnosed with colon cancer between 1998 and 2019, ≥ 65 years old, and at least two HRQOL surveys were identified and grouped. Group 1 comprised patients with an initial survey pre‐diagnosis and a follow‐up survey within 1 year from diagnosis, while Group 2 included patients with an initial survey within 1 year from diagnosis and a follow‐up survey 2 years later. Bayesian linear regression identified predictors of physical (PCS) and mental (MCS) score changes, incorporating demographic, socioeconomic, comorbidity, and cancer‐related variables. Results: The analysis included 1139 patients (Group 1: 571; Group 2: 568). PCS and MCS scores decreased substantially at diagnosis (4.4 and 2.2 points, respectively) compared to baseline but changed minimally from baseline (−0.3 and +0.6 points, respectively) 2 years post‐diagnosis. Advanced disease stage and baseline ADL limitations were the strongest predictors of HRQOL decline, both from pre‐ to post‐diagnosis and from diagnosis to 2 years later. Conversely, positive baseline health perception was protective (probability of direction > 90% for all estimates). Conclusions: Among colon cancer patients, patient‐reported baseline health perception and functional limitations emerged as critical determinants of HRQOL trajectories, outweighing socioeconomic and treatment‐related factors. These factors remain consistent upon diagnosis and during survivorship. Assessment of these factors could estimate HRQOL outcomes and facilitate early identification of at‐risk patients.
- 17 June 2026
- Published by Wiley in Cancer Medicine
- Vol. 15 (6) , e72043
Open AccessNationwide Analysis of Sociodemographic Disparities in Health‐Related Quality of Life of Colon Cancer Survivors
Background: Health‐related quality of life (HRQOL) measures patients' overall well‐being and can help guide survivorship care. Evidence comparing individual patients' HRQOL pre‐ and post‐diagnosis is limited due to the unpredictable nature of cancer onset. Our study aims to evaluate trends in HRQOL in elderly colon cancer patients and identify predictors of HRQOL change across two key transitions: (1) pre‐ to post‐diagnosis, and (2) at diagnosis versus 2 years later, with the goal of identifying at‐risk patients for a significant HRQOL decrease. Methods: Utilizing the SEER‐Medicare Health Outcomes Survey (SEER‐MHOS), patients diagnosed with colon cancer between 1998 and 2019, ≥ 65 years old, and at least two HRQOL surveys were identified and grouped. Group 1 comprised patients with an initial survey pre‐diagnosis and a follow‐up survey within 1 year from diagnosis, while Group 2 included patients with an initial survey within 1 year from diagnosis and a follow‐up survey 2 years later. Bayesian linear regression identified predictors of physical (PCS) and mental (MCS) score changes, incorporating demographic, socioeconomic, comorbidity, and cancer‐related variables. Results: The analysis included 1139 patients (Group 1: 571; Group 2: 568). PCS and MCS scores decreased substantially at diagnosis (4.4 and 2.2 points, respectively) compared to baseline but changed minimally from baseline (−0.3 and +0.6 points, respectively) 2 years post‐diagnosis. Advanced disease stage and baseline ADL limitations were the strongest predictors of HRQOL decline, both from pre‐ to post‐diagnosis and from diagnosis to 2 years later. Conversely, positive baseline health perception was protective (probability of direction > 90% for all estimates). Conclusions: Among colon cancer patients, patient‐reported baseline health perception and functional limitations emerged as critical determinants of HRQOL trajectories, outweighing socioeconomic and treatment‐related factors. These factors remain consistent upon diagnosis and during survivorship. Assessment of these factors could estimate HRQOL outcomes and facilitate early identification of at‐risk patients.
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure 2 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Figure 2 shows posterior predictive checks for the model predicting PCS change from diagnosis to two years. The three panels compare observed versus replicated data in terms of overall distribution, mean, and variance to assess model fit.
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure 2 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Figure 2 shows posterior predictive checks for the model predicting PCS change from diagnosis to two years. The three panels compare observed versus replicated data in terms of overall distribution, mean, and variance to assess model fit.
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure 1 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Figure 1 shows violin plots with medians and interquartile ranges for baseline physical (PCS) and mental (MCS) component scores, before versus after diagnosis (Panel A), and at diagnosis versus two years later (Panel B).
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure 1 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Figure 1 shows violin plots with medians and interquartile ranges for baseline physical (PCS) and mental (MCS) component scores, before versus after diagnosis (Panel A), and at diagnosis versus two years later (Panel B).
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Data from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Background:Evidence comparing patients’ health-related quality of life (HRQOL) before and after breast cancer diagnosis, and factors influencing these trajectories, is limited because of unpredictable onset. We analyzed predictors of HRQOL change in elderly patients with breast cancer across two transitions: pre- to post-diagnosis and at diagnosis versus 2 years later.Methods:Using Surveillance, Epidemiology, and End Results–Medicare Health Outcomes Survey linkage, we analyzed patients with breast cancer >65 years who completed two HRQOL surveys. Two cohorts were examined: patients with surveys before and within 1 year after diagnosis (cohort 1) and those with surveys within 1 year of diagnosis and 2 years after diagnosis (cohort 2). Bayesian regression identified predictors of physical component summary (PCS) and mental component summary (MCS) score changes.Results:In cohort 1 (n = 1,546), advanced stage was associated with greater HRQOL decline, with mean PCS changes ranging from −2.1 to −6.5 and MCS from −2.0 to −5.1. Baseline activities of daily living (ADL) limitations were associated with mean PCS declines ranging from −2.0 to −2.6 and mean MCS decline of −2.0. Better baseline health perception was protective with mean PCS increases ranging from +1.4 to +3.4. In cohort 2 (n = 891), baseline ADL limitations were associated with mean PCS declines ranging from −1.6 to −5.5 and mean MCS decline of −2.5, whereas poor health perception was associated with mean PCS decline of −3.2. Excellent health perception was protective (mean MCS: +4.1). Surgery, chemotherapy, and radiation had no HRQOL affect.Conclusions:Baseline health perception and ADL limitations are critical determinants of HRQOL trajectories, outweighing treatments.Impact:Early intervention strategies based on baseline assessments may improve survivorship.
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessData from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Background:Evidence comparing patients’ health-related quality of life (HRQOL) before and after breast cancer diagnosis, and factors influencing these trajectories, is limited because of unpredictable onset. We analyzed predictors of HRQOL change in elderly patients with breast cancer across two transitions: pre- to post-diagnosis and at diagnosis versus 2 years later.Methods:Using Surveillance, Epidemiology, and End Results–Medicare Health Outcomes Survey linkage, we analyzed patients with breast cancer >65 years who completed two HRQOL surveys. Two cohorts were examined: patients with surveys before and within 1 year after diagnosis (cohort 1) and those with surveys within 1 year of diagnosis and 2 years after diagnosis (cohort 2). Bayesian regression identified predictors of physical component summary (PCS) and mental component summary (MCS) score changes.Results:In cohort 1 (n = 1,546), advanced stage was associated with greater HRQOL decline, with mean PCS changes ranging from −2.1 to −6.5 and MCS from −2.0 to −5.1. Baseline activities of daily living (ADL) limitations were associated with mean PCS declines ranging from −2.0 to −2.6 and mean MCS decline of −2.0. Better baseline health perception was protective with mean PCS increases ranging from +1.4 to +3.4. In cohort 2 (n = 891), baseline ADL limitations were associated with mean PCS declines ranging from −1.6 to −5.5 and mean MCS decline of −2.5, whereas poor health perception was associated with mean PCS decline of −3.2. Excellent health perception was protective (mean MCS: +4.1). Surgery, chemotherapy, and radiation had no HRQOL affect.Conclusions:Baseline health perception and ADL limitations are critical determinants of HRQOL trajectories, outweighing treatments.Impact:Early intervention strategies based on baseline assessments may improve survivorship.
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Table 2 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table S2 presents the posterior estimates, 95% credible intervals, and probabilities of direction (PD) for each predictor in the Bayesian regression model evaluating the association between patient factors and changes in Mental Component Summary (MCS) scores between pre-diagnosis and diagnosis relative to the minimally clinically important difference (MCID) threshold.
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table 2 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table S2 presents the posterior estimates, 95% credible intervals, and probabilities of direction (PD) for each predictor in the Bayesian regression model evaluating the association between patient factors and changes in Mental Component Summary (MCS) scores between pre-diagnosis and diagnosis relative to the minimally clinically important difference (MCID) threshold.
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Table 4 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table S4 compares demographic, socioeconomic, clinical, and health-related baseline characteristics between two groups of elderly breast cancer patients from the SEER-MHOS dataset: those who completed both pre- and post-diagnosis HRQOL surveys (n=1,546) and those who completed only a pre-diagnosis survey (n=1,416). Differences are presented as proportions or mean differences with corresponding 95% confidence intervals. The table highlights potential selection biases in survey completion and the representativeness of the longitudinal cohort
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table 4 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table S4 compares demographic, socioeconomic, clinical, and health-related baseline characteristics between two groups of elderly breast cancer patients from the SEER-MHOS dataset: those who completed both pre- and post-diagnosis HRQOL surveys (n=1,546) and those who completed only a pre-diagnosis survey (n=1,416). Differences are presented as proportions or mean differences with corresponding 95% confidence intervals. The table highlights potential selection biases in survey completion and the representativeness of the longitudinal cohort
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Table 5 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table S5 compares the baseline demographic, clinical, and health-related characteristics of breast cancer patients who completed both the diagnosis and 2-year follow-up surveys versus those who only completed the diagnosis survey, highlighting potential selection bias in longitudinal HRQOL analysis.
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table 5 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table S5 compares the baseline demographic, clinical, and health-related characteristics of breast cancer patients who completed both the diagnosis and 2-year follow-up surveys versus those who only completed the diagnosis survey, highlighting potential selection bias in longitudinal HRQOL analysis.
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Table 3 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table 3 presents the R-hat convergence diagnostics for all Bayesian models used in the study. It includes the minimum and maximum R̂ values for each model across both cohorts and outcome types (PCS and MCS). All R̂ values are close to 1, indicating excellent convergence of the Markov chains.
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table 3 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Table 3 presents the R-hat convergence diagnostics for all Bayesian models used in the study. It includes the minimum and maximum R̂ values for each model across both cohorts and outcome types (PCS and MCS). All R̂ values are close to 1, indicating excellent convergence of the Markov chains.
- Open Access
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure 3 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Figure 3 shows posterior predictive checks for the model predicting MCS change from diagnosis to two years. The figure includes kernel density overlays and distributions of replicated vs observed means and variances.
- 9 January 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure 3 from Predicting Health-Related Quality of Life Trajectories in Elderly Patients with Breast Cancer: From Prediagnosis to Long-term Follow-up
Supplementary Figure 3 shows posterior predictive checks for the model predicting MCS change from diagnosis to two years. The figure includes kernel density overlays and distributions of replicated vs observed means and variances.