Does Risk of Treatment Induced Cardiotoxicity Influence Cancer Care? An Assessment of Cancer Directed Treatment Planning, Delivery, and Cardiac Outcomes Among Breast Cancer Patients

Abstract

Treatment with radiation therapy confers the benefit of lower rates of recurrence and longer survival. However, treatment comes with risk of short and long-term adverse effects. Cardiotoxicity is an established risk of treatment, particularly for thoracic tumors such as breast cancer. Such risks may deter patients from accepting treatment that includes radiation therapy. This dissertation uses novel methods to explore choices made during 1) treatment planning, including use of new simulation techniques, 2) treatment delivery, with a focus on radiation therapy delivery, and examines resulting 3) cardiovascular outcomes among a population-based sample of women with breast cancer. The studies included in this dissertation use the Surveillance, Epidemiology, and End Results (SEER) program data linked with Medicare claims (SEER-Medicare). The study population consists of women aged 66 and older who were newly diagnosed with breast cancer between 2000 and 2017. Claims-based research methods were used (i.e., identification of services through use of billing codes [ICD-9-CM/ICD-10 CM, CPT, HCPCS, etc.]) to measure radiation treatment planning services, cancer-directed treatment received, and to identify clinical outcomes of interest. Respiratory motion management simulation (RMMS) is a newer simulation technique used to understand the full impacts of the breath cycle on target volumes of treatment and exposure to nearby organs. We found that while most patients who receive radiation therapy undergo complex planning (93.0%) and some form of simulation (95.0%), very few received RMMS (3.4%) and there was very little uptake over the study period (overall use of 2.2% in 2014 increasing to 5.8% in 2017). Use of RMMS also varied across key patient and provider characteristics, with the largest differences observed across tumor laterality (left: 6.1% vs right: 0.7%) and rural/urban status (Metro: 3.7% vs. non-metro, adjacent: 1.9% vs. non-metro, non-adjacent: 1.6%). Higher uptake over time was also observed among women with left sided tumors (3.6% in 2014 to 10.3% in 2017, p<0.001) compared to right sided tumors, whose rates were very steady over the study period (0.7% in 2014 to 1.1% in 2017, p = 0.03). Women were less likely to see a provider who used RMMS if they sought care at smaller, rural institutions (smaller size: OR = 0.747 [95% CI: 0.596-0.936]; rural: OR = 0.721 [95% CI: 0.622-0.836]). After controlling for other relevant factors, women with left sided tumors were found to be 9 times more likely to receive RMMS compared to women with right sided tumors (OR = 0.9001 [95% CI: 7.121-11.376]) suggesting there is increased concern for tumors located near the heart. We also found that women with left sided tumors were slightly less likely to receive radiation therapy treatment (OR = 0.964 (0.946-0.982), p < 0.0001), as are those with underlying cardiac comorbidities (OR = 0.971 (0.951-0.991), p = 0.0051). Women with underlying cardiac comorbidities were also less likely to receive chemotherapy (20.4% vs. 25.8%, p < 0.0001) and more likely to receive mastectomy (40.6% vs. 38.8%, p < 0.0001) suggesting a pattern of choosing heart sparing techniques for treatment. Significantly fewer women with left sided cancers (78.6% vs. 79.2%, p = 0.0012) and underlying cardiac comorbidities (76.4% vs. 80.5%, p < 0.0001) received guideline concordant care. This study demonstrated a further signal that treatment choice may be dependent on concerns of the associated cardiac risk of treatment. Finally, we found that women with left sided tumors were not more likely than women with right sided tumors to experience long term cardiac morbidity and mortality events. Women with left sided tumors had a higher crude incidence of myocardial infarction (IRR=1.026 [95% CI: 1.001-1.052], p < 0.05) and coronary artery disease (IRR=1.027 [95% CI: 1.012-1.043], p < 0.001). However, these findings were no longer significant in adjusted analyses. In fact, receipt of radiation therapy was associated with a significantly lower risk of cardiac-related mortality (HR: 0.725 [95% CI: 0.699-0.751], p<0.0001), reaffirming the improvements in the effectiveness and safety of radiation treatment in recent years. The cumulative findings of this dissertation suggest that there is a signal for some concern regarding cardiotoxic effects of treatment. This was observed both in the planning and treatment phases of the breast cancer care continuum. However, the rates of new simulation techniques to better target treatment were quite low. Efforts may be needed to promote the adoption of new planning techniques that could result in safer and more precise delivery of treatment. These findings also showed no evidence of excess risk due to exposure near the heart (i.e., left side exposure compared to right side exposure), supporting the notion that long term cardiac risk following treatment with radiation therapy has greatly improved. Future qualitative studies could further characterize the decision-making process, both at the patient and the provider level, to better understand how the perceived risks of treatment influence the cancer-directed treatment modality chosen. Shared decision-making approaches can also ensure that the most up to date evidence on risks and safety of treatment are presented to the patient during the treatment planning process, leading to more individualized and effective care for women with breast cancer.