Date of Award

2025

Degree Type

Dissertation

Degree Name

Doctor of Philosophy in Pharmaceutical Sciences

Department

Biomedical and Pharmaceutical Sciences

First Advisor

Ami Vyas

Abstract

Each year in the United States, more individuals are diagnosed with skin cancer than all other forms of cancer combined. About 20 percent of Americans will be diagnosed with skin cancer by the time they are 70 years old. While only one percent skin cancer is diagnosed as invasive melanoma, it accounts for the majority of skin cancer deaths. When diagnosed and treated appropriately, cutaneous melanoma has a high survival rate, yet established, evidence based, clinical guidelines are not always adhered to and subsequent patient outcomes are impacted. In addition to deviation from established clinical guidelines, newly introduced medications contribute to staggering costs to guideline concordant treatment and ultimately, substantial costs are born by the US healthcare system. The purpose of this dissertation was to investigate disparities in provision of guideline concordant treatment at patient level, examine the relationship between guideline concordant treatment and overall mortality and melanoma specific mortality, and healthcare resource utilization, and the trends in costs attributed to the treatments available for primary treatment of cutaneous melanoma.

Manuscript 1: The objective of this study was to identify disparities in guideline-concordant primary treatment among older Medicare fee-for-service beneficiaries diagnosed with stage III or stage IV cutaneous melanoma. We conducted a retrospective cohort study using the Surveillance, Epidemiology and End Results (SEER)-Medicare linked database. We included all adults older than 65 years, who had a new, first-time diagnosis for cutaneous melanoma from 2011-2015, and no history of previous cancer diagnosis. Guideline-concordant primary treatment was defined as receipt of primary cancer treatment during the first six months after diagnosis, according to the published National Comprehensive Cancer Network (NCCN) guidelines. We identified the significant differences between those who received guideline-concordant primary treatment versus those who failed to receive guideline-concordant primary treatment and conducted multivariable logistic regression to identify predictors of failure to receive guideline-concordant primary treatment. We also characterized treatment regimens in patients who received guideline-concordant primary treatment. The study population consisted of 1,467 adults; 18.9% of our study population failed to receive guideline-concordant primary treatment, while 81.1% received guideline-concordant primary treatment. In the final reduced model, the adjusted odds ratio (AOR) for failure to receive guideline-concordant primary treatment increased when the following factors were present: diagnosis at stage IV cancer (compared to stage III, local, AOR 1.597, 95% Confidence Interval (CI) 1.077-2.369, p 0.0199), having a Charlson comorbidity index of 0 (compared to two or more, AOR 1.543, 95% CI 1.108-2.150, p 0.0103), having good performance status (compared to poor performance status, AOR 1.401, 95% CI 1.007-1.949, p < 0.0426), relatively older age (85 years and older compared to 66-75 years old, AOR 2.522, 95% CI 1.710-3.722, p < 0.0001), being non-White (compared to white, AOR 2.674, 95% CI 1.309-5.352, p 0.0067), not having a spouse/partner (compared to having a partner, AOR 1.536, 95% CI 1.143-2.065, p 0.0045), and living in the Western region of the United States (compared to living in the Northeast region, AOR 2.001, 95% CI 1.296-3.091, p 0.0018). For localized Stage 3 tumor, of the 232 individuals who received guideline-concordant primary treatment, 84.9% received lymph node dissection and wide excision of the primary tumor; 15.1% received lymph node dissection only. For Stage 3, in transit tumor, of the 615 who received guideline-concordant primary treatment, 52.5% had the primary tumor surgically excised only, 5.2% had the primary tumor surgically excised and received systemic therapy, 9.3% had the primary tumor surgically excised and received radiation therapy, 12.4% had the primary tumor surgically excised and received an intralesional injection, and 2.1% had the primary tumor surgically excised and received additional treatment combinations. Additionally, 7.3% received systemic therapy and local ablation therapy, 9.4% received systemic therapy and other treatment combinations. Of the 343 individuals diagnosed with stage IV melanoma who received guideline-concordant primary treatment, the majority received surgical excision and additional treatment (systemic therapy (6.4%) and radiation therapy (39.4%), or systemic therapy and other additional treatments (12.0%) or radiation therapy (15.2%) and other additional treatments (9.0%), or palliative care (8.7%), or observation (18.4%)), 7.3% received radiation only, and 18.1% received other treatment combinations which aligned with NCCN guidelines. Our findings suggest that stage IV tumor diagnosis, having no comorbidities and good performance status, relatively older age, not having a partner, and non-White race increased odds of failure to receive guideline-concordant primary treatment for cutaneous melanoma in Medicare fee-for-service beneficiaries. Support systems and regular interaction with the health care system can help enhance the receipt of guideline-concordant primary treatment in patients with advanced cutaneous melanoma.

Manuscript 2: We examined the association between guideline-concordant primary treatment on healthcare resource utilization and clinical outcomes, among older Medicare fee-for-service beneficiaries diagnosed with stage III or IV cutaneous melanoma. We conducted a retrospective observational cohort study using the Surveillance, Epidemiology and End Results (SEER)-Medicare linked database. We included adults older than 65 years who had a new, first-time diagnosis for cutaneous melanoma from 2011-2015, and had no history of a previous cancer diagnosis. We defined guideline-concordant primary treatment as receipt of primary cancer treatment during the first six months after diagnosis, according to National Comprehensive Cancer Network (NCCN) guidelines. We examined clinical outcomes by assessing all-cause mortality and melanoma-specific mortality, and healthcare resource utilization including emergency department visits, inpatient hospitalizations, outpatient visits, and office-based physician visits. Generalized linear regressions and Cox proportional hazards model were conducted to examine the association between guideline-concordant primary treatment and healthcare resource utilization and mortality, respectively. A log offset was applied to each model to account for different measurement periods for each patient. Propensity score-matched (PSM) analyses for all outcomes were conducted to reduce selection bias. In our original cohort, of 1,467 adults included, 81.1% received guideline-concordant primary treatment. Patients who failed to receive guideline-concordant primary treatment had higher hazards of all-cause mortality (adjusted hazard ratio (AHR) from a reduced model 1.747, 95% confidence interval (CI) 1.457-2.094, p < 0.0001) and melanoma-specific mortality (AHR 1.644, 95% CI 1.329-2.033, p < 0.0001) compared to those who received guideline-concordant primary treatment. Rates of all-cause inpatient hospitalizations were 79% higher for those who failed to receive guideline-concordant primary treatment (adjusted incident rate ratio (AIRR) 1.787, 95% CI 1.306-2.444 p =0.0003), but 31% lower for all-cause outpatient visits (AIRR 0.691, 95% CI 0.604-0.789, p < 0.0001), 67% lower for melanoma-specific outpatient visits (AIRR 0.330, 95% CI 0.221-0.491, p < 0.0001), 17% lower for all-cause office visits (AIRR 0.836, 95% CI 0.739-0.945, p= 0.0042), and 48% for melanoma-specific office visits (AIRR 0.521, 95% CI 0.368-0.737, p 0.0002) compared to those who received guideline-concordant primary treatment. The findings remained consistent for PSM analysis. Our findings suggest that failure to receive guideline-concordant primary treatment for advanced cutaneous melanoma is significantly associated with increased all-cause and melanoma-specific mortality. Patients receiving guideline-concordant primary treatment utilized office-based physicians and outpatient visits more but were less likely to be hospitalized. Examining clinical outcomes and healthcare resource utilization based on receipt of any guideline-concordant primary treatment recommendation reinforces the important role of evidence-based published guidelines.

Manuscript 3: We described the trends of use and cost of the melanoma-specific treatment from 2011 - 2018, after initiation of first-line systemic therapy for cutaneous melanoma in melanoma patients. We conducted a retrospective observational study using administrative claims data. Patients 18 years and older, who initiated a first-line systemic therapy for cutaneous melanoma from January 1, 2011 - June 30, 2018, and had claims for cutaneous melanoma diagnosis in the previous 12 months were included. Melanoma treatment specific costs were calculated and aggregated to compute average six-month costs which were stratified by patient-level characteristics. Average monthly cost ratios were evaluated by building generalized linear models with log link and gamma distribution. We also assessed trends in treatments used during the same period. In the final reduced adjusted model of per treated member per month (PTMPM) costs, we observed significant cost increases each year compared to the year 2011. For average monthly costs 6-months after initiation of treatment, compared to 2011, the PTMPM cost increases were : 2012 (288%, 95% CI 184%-432%, p < 0.0001), 2013 (481%, 95% CI 327%-692%, p < 0.0001), 2014 (373%, 95% CI 250%-538%, p < 0.0001), 2015 (331%, 95% CI 223%-475%, p < 0.0001), 2016 (318%, 95% CI 219%-446%, p < 0.0001), 2017 (443%, 95% CI 318%-606%, p < 0.0001), and 2018 (474%, 95% CI 342%-644%, p < 0.0001). The cost increases aligned with the introduction of new therapies. In 2011, the five most used systemic therapies in our cohort were carboplatin (21.3%), interferon (16.7%), cisplatin (16.7%), paclitaxel (16.1%), and ipilimumab (6.9%). The top four used systemic therapies, which were traditional treatments options, had PTMPM costs ranging from $221 to $981, compared to the ipilimumab which had a PTMPM cost of $31,917. By 2018, nivolumab accounted for 51.0% of the systemic therapies used that year, followed by pembrolizumab (21.6%), carboplatin (6.4%), cisplatin (4.8%), and paclitaxel (4.3%). The most used systemic therapy in 2012 (carboplatin) cost on average $221 per month for 6 months, compared to $18,611 for the most used systemic therapy in 2018 (nivolumab). The introduction of newer therapies, which appear to be preferred by prescribers and patients, has increased the overall cost of systemic therapies, year over year. The melanoma treatment landscape continues to evolve, newly approved indications for previously approved systemic therapies will continue to drive costs.

Download

Available for download on Monday, May 10, 2027

Share

COinS