Kelly Carter Nelson

Background: Melanoma, a highly aggressive form of skin cancer, is the second most common type of cancer for adolescent and young adult (AYA, ages 15-39 years) patients. AYA patients with melanoma may turn to internet sources, especially artificial intelligence (AI) chatbots, to manage uncertainty about prognosis and treatment. Objective: This study aims to evaluate the quality, empathy, and readability of responses generated by leading AI chatbots when addressing the top unmet needs of AYA patients with melanoma receiving treatment. Methods: Our research team recently surveyed 152 AYA patients with melanoma using the Needs Assessment Service Bridge, a validated instrument that assesses psychosocial needs for AYA patients with cancer. The survey identified the top 5 needs for advanced AYA patients with melanoma receiving treatment. Each need was reframed into a question and brief clinical history, then entered into each chatbot by 5 individuals who cleared their prequestion and postquestion history. Chatbot responses were evaluated to assess information quality (Global Quality Score [GQS] and DISCERN), accessibility and readability (GQS, Flesch Kincaid Grade Level, Flesch Reading Ease), and perceived empathy (Perceived Empathy of Technology Scale [PETS], including domains of Emotional Responsiveness [PETS-ER], Understanding and Trust [PETS-UT]). Results: Across 75 chatbot responses, ChatGPT achieved the highest average quality (mean GQS 4.42, SD 0.32; mean DISCERN 3.24, SD 0.31) and empathy (mean PETS-ER 5.35, SD 1.85; mean PETS-UT 6.36, SD 1.83), though with greater variability. Copilot produced the lowest quality and empathy scores, while Gemini responses were consistently midrange. PETS-UT exceeded PETS-ER across all models, suggesting stronger cognitive empathy than emotional responsiveness. Readability analysis showed outputs exceeded the average US reading level (mean Flesch Kincaid Grade Level 11.82, SD 1.44; mean FRE 38.60, SD 9.00), limiting accessibility. The most readable responses were found in question 2, which also scored higher in quality and empathy, whereas questions 4 and 5 produced the most complex, difficult-to-read responses corresponding with lower quality and empathy ratings. Conclusions: AI chatbots can provide moderately accurate and supportive responses to needs of AYA patients with melanoma, but outputs are inconsistent, written above the recommended reading level for health information, and limited in empathy. Question framing strongly influenced chatbot performance, with more emotional prompts drawing greater empathy, and readability aligning with both quality and empathy. Chatbot use in this population should remain adjunctive, with further research needed to standardize quality, improve readability, and enhance empathetic communication.

Melanoma is projected to become the second most diagnosed cancer in the United States by 2040, emphasizing the importance of early detection for improving survival outcomes. Accurately estimating late-stage melanoma incidence at granular geographical levels is challenging due to its rarity and data sparsity. Instead of relying on aggregated reported health statistics on melanoma cases, which are commonly used in cancer surveillance research, we utilize the incidence-based individual case database, and propose a Bayesian Heterogeneous Spatio-Temporal Logistic Regression with Nonlinear Demographic Effect (BHSTLR-NDE) model to provide county-level annual estimates of late-stage melanoma incidence in Texas from 2000 to 2018, using data obtained through a request to the Texas Cancer Registry. The BHSTLR-NDE integrates county-level covariates, flexible case-level demographic effects, and a reduced-rank spatial modelling approach to effectively address data sparsity, spatio-temporal non-stationarity, and demographic heterogeneity. The superiority of the proposed BHSTLR-NDE over competing models is demonstrated through extensive simulation studies and an application to melanoma registry data. Results indicate that BHSTLR-NDE effectively borrows information across spatial, temporal, and demographic dimensions, improving estimation accuracy and robustness. These findings highlight the utility of the proposed modelling approach in enhancing cancer surveillance and guiding targeted interventions.

Adolescents and young adults (AYAs, age 15-39 years) with melanoma have needs unique from pediatric or adult populations due to interference with developmental milestones. This study aims to assess the needs of AYA melanoma patients using the Needs Assessment and Service Bridge (NA-SB) questionnaire. A REDCap survey based on the AYA NA-SB was distributed via email, social media, newsletters, and conferences between August to October 2024. Data were reported for the overall population and analyzed by sex, treatment phase, and exposure to immune checkpoint blockers (ICBs). One hundred fifty-two patients participated in the survey. Male and female AYAs had different need profiles, with the males expressing greater need in understanding the role of genetics in diagnosis (51.5% vs. 33.9%, p = 0.019) and the sexual side effects of treatment (38.3% vs. 16.6%, p = 0.016) and the females reporting greater needs in understanding their disease status (71.3% vs. 48.5%, p = 0.003). AYAs on active treatment reported greater needs than those who completed treatment in managing work and life during treatment (55.5% vs. 27.1%, p = 0.03), being equipped to cope with diagnosis (42.8% vs. 25%, p = 0.029), and managing the emotional side effects of treatment (62% vs. 29.5%, p = 0.002). AYAs treated with ICBs had greater needs in managing long-term side effects of treatment (70.8% vs. 31.6%, p = 0.0005), pain management (30.6% vs. 11.6%, p = 0.01), and the physical side effects of treatment (36.7% vs. 16.5%, p = 0.021). AYA melanoma patients express unique needs based on sex, treatment phase, and type of systemic therapy. Assessing patient needs and connecting AYAs to relevant supportive services may help mitigate unique challenges encountered during melanoma treatment and in post-treatment survivorship.

BACKGROUND Melanoma, a highly aggressive form of skin cancer, is the second most common type of cancer for adolescent and young adult (AYA, ages 15-39 years) patients. AYA melanoma patients may turn to internet sources, especially AI-chatbots, to manage uncertainty about prognosis and treatment. OBJECTIVE To evaluate the quality, empathy, and readability of responses generated by leading AI chatbots when addressing the top unmet needs of AYA melanoma patients receiving treatment. METHODS Our research team recently surveyed 152 AYA melanoma patients using the Needs Assessment Service Bridge (NA-SB), a validated instrument that assesses psychosocial needs for AYA cancer patients. The survey identified the top 5 needs for advanced AYA melanoma patients receiving treatment. Each need was reframed into a question and brief clinical history, then entered into each chatbot by five individuals who cleared their pre- and post-question history. Chatbot responses were evaluated to assess information quality (Global Quality Score (GQS) and DISCERN), accessibility and readability (GQS, Flesch Kincaid Grade Level, Flesch Reading Ease), and perceived empathy (Perceived Empathy of Technology (PETS)). RESULTS Across 75 chatbot responses, ChatGPT achieved the highest average quality (mean GQS 4.4, mean DISC 3.2) and empathy (PETS-ER 5.4, PETS-UT 6.4), though with greater variability (SD~1.8). Copilot produced the lowest quality and empathy scores, while Gemini responses were consistently midrange. PETS-UT exceeded PETS-ER across all models, suggesting stronger cognitive empathy than emotional responsiveness. Readability analysis showed outputs exceeded the average U.S. reading level (mean FKGL 11.8, FRE 38.6), limiting accessibility. The most readable responses were found in Question 2, which also scored higher in quality and empathy, whereas Question 4 and 5 produced the most complex, difficult to read responses corresponding with lower quality and empathy ratings. CONCLUSIONS AI chatbots can provide moderately accurate and supportive responses to AYA melanoma patient needs, but outputs are inconsistent, written above the recommended reading level for health information, and limited in empathy. Question framing strongly influenced chatbot performance, with more emotional prompts drawing greater empathy, and readability aligning with both quality and empathy. Chatbot use in this population should remain adjunctive, with further research needed to standardize quality, improve readability, and enhance empathetic communication.

Introduction: School policy can encourage sun safe habits, such as wearing hats and applying sunscreen. However, sun safety policies (SSP) have not been formally assessed for Texas independent school districts (ISDs), particularly in counties with the highest melanoma incidence relative risk (RR). This study aims to assess the presence, strength, and intent of SSPs across Texas ISDs located in counties with the highest and lowest melanoma incidence. We also identify factors correlated with stronger SSP. Methods: Eleven components of SSPs from 102 ISDs were evaluated in this cross-sectional study by examining school district websites, official documents, social media, media appearances, statements by school officials, and the Texas Education Agency’s online database. Coders were trained to score each policy’s content, presence, and strength. Results: Policies for sunscreen use and hats existed in 94% (n = 96) and 92% (n = 94) of ISDs, respectively. In counties with the highest melanoma incidence RR, 30% (n = 15) and 44% (n = 22) of ISDs allocated resources for sun safety and outdoor shade, compared to 2% (n = 1) and 3% (n = 2) in low-risk counties. No ISDs had SSPs on UV protective clothing, accountability, or modeling sun safety behaviors. SSP strength was positively correlated with percentage of school nurses (ρ = 0.564, P < 0.001), community median household income (ρ = 0.431, P < 0.001), percentage of female students (ρ = 0.461, P < 0.001), and tax rate (ρ = 0.366, P = 0.0002). Negative correlations were found with percentage of central staff administration (ρ = −0.523, P < 0.001) and graduation rates (ρ = −0.335, P < 0.001). Conclusion: Our findings underscore the need for interventions to strengthen SSPs across Texas.

Background: Melanoma currently ranks as the fifth leading cancer diagnosis and is projected to become the second most common cancer in the United States by 2040. Melanoma detected at earlier stages may be treated with less-risky and less-costly therapeutic options. Objective: This study aims to analyze temporal and spatial trends in melanoma incidence by stage at diagnosis (overall, early, and late) in Texas from 2000 to 2018, focusing on demographic and geographic variations to identify high-risk populations and regions for targeted prevention efforts. Methods: We used melanoma incidence data from all 254 Texas counties from the Texas Cancer Registry (TCR) from 2000 to 2018, aggregated by county and year. Among these, 250 counties reported melanoma cases during the period. Counties with no cases reported in a certain year were treated as having no cases. Melanoma cases were classified by SEER Summary Stage and stratified by the following four key covariates: age, sex, race and ethnicity, and stage at diagnosis. Incidence rates (IRs) were calculated per 100,000 population, and temporal trends were analyzed using joinpoint regression to determine average annual percentage changes (AAPCs) with 95% CIs for the whole time period (2000‐2018), the most recent 10-year period (2009‐2018), and the most recent 5-year period (2014‐2018). Heat map visualizations were developed to assess temporal trends by patient age, year of diagnosis, stage at diagnosis, sex, and race and ethnicity. Spatial cluster analysis was conducted using Getis-Ord Gi* statistics to identify county-level geographic clusters of high and low melanoma incidence by stage at diagnosis. Results: A total of 82,462 melanoma cases were recorded, of which 74.7% (n=61,588) were early stage, 11.3% (n=9,352) were late stage, and 14% (n=11,522) were of unknown stage. Most cases were identified as males and non-Hispanic White individuals. Melanoma IRs increased from 2000 to 2018, particularly among older adults (60+ years; AAPC range 1.20%-1.84%; all P values were <.001), males (AAPC 1.59%; P<.001), and non-Hispanic White individuals (AAPC of 3.24% for early stage and 2.38% for late stage; P<.001 for early stage and P = .03 for late state). Early-stage diagnoses increased while the rates of late-stage diagnoses remained stable for the overall population. The spatial analysis showed that urban areas had higher early-stage incidence rates (P=.06), whereas rural areas showed higher late-stage incidence rates (P=.05), indicating possible geographic-based differences in access to dermatologic care. Conclusions: Melanoma incidence in Texas increased over the study time period, with the most-at-risk populations being non-Hispanic White individuals, males, and individuals aged 50 years and older. The stable rates of late-stage melanoma among racial and ethnic minority populations and rural populations highlight potential differences in access to diagnostic care. Future prevention efforts may benefit from increasing access to dermatologic care in areas with higher rates of late-stage melanoma at diagnosis.

Desmoplastic melanoma (DM) is an uncommon subtype of cutaneous melanoma that presents distinct diagnostic and treatment challenges. This review aims to explore the role of adjuvant radiation therapy (RT) in managing DM. To evaluate this question, we reviewed relevant published reports on DM and its treatment and synthesized these findings. It was found that the clinical behavior of DM varies significantly based on its classification as either “pure” DM (pDM, ≥90% desmoplastic features) or mixed DM (mDM, ≤90% desmoplastic features). Patients with pDM have a uniquely high risk of local recurrence but a relatively lower likelihood of nodal disease. Recent studies question the necessity of sentinel lymph node biopsy for pDM patients while illustrating impressive response rates to immune checkpoint inhibition. Most data supporting adjuvant RT predate these changes in surgical management and systemic therapy, yet consistently demonstrate that adjuvant RT reduces the absolute risk of local recurrence by >50%, without significant long-term toxicity. Thus, the existing literature continues to support the conclusion that adjuvant RT effectively reduces the likelihood of local recurrence in pDM patients. Although evolving surgical and systemic therapies are reshaping treatment approaches, adjuvant RT should remain a standard of care.