Ross Harrison

Body mass index and attitudes towards health behaviors among women with endometrial cancer before and after treatment

Some experts have argued that obesity-related malignancies such as endometrial cancer are a "teachable moment" that lead to meaningful changes in health behaviors. It is unclear if endometrial cancer survivors lose weight following treatment. Our goal with this investigation was to evaluate post-treatment changes in body mass index (BMI) and attitudes towards health behaviors in endometrial cancer survivors. Incident endometrial cancer cases undergoing surgery between 2009-2015 were identified in the Marketscan Commercial database and linked with BMI data and health behavior questionnaires from the Marketscan Health Risk Assessment database. Patients were excluded for insufficient BMI data. Standard statistical methods, including the two-sample Wilcoxon rank sum test, χ 655 patients with a median age of 54 (IQR 49-58) were identified and analyzed. Median duration of follow-up was 595 days (IQR 360-1091). Mean pre- and post-treatment BMI was 35.5 kg/m Most endometrial cancer survivors gain weight or maintain the same weight following treatment. No post-treatment changes in attitudes regarding weight-related behaviors were observed. The systematic delivery of evidence-based weight loss interventions should be a priority for survivors of endometrial cancer.

Beyond post-operative readmissions: analysis of the impact of unplanned readmissions during primary treatment of advanced-stage epithelial ovarian cancer on long-term oncology outcome

Multiple studies have assessed post-operative readmissions in advanced ovarian cancer. To evaluate all unplanned readmissions during the primary treatment period of advanced epithelial ovarian cancer, and the impact of readmission on progression-free survival. This was a single institution retrospective study from January 2008 to October 2018. Χ A total of 484 patients (279 primary cytoreductive surgery, 205 neoadjuvant chemotherapy) were analyzed. In total, 272 of 484 (56%; 37% primary cytoreductive surgery, 32% neoadjuvant chemotherapy, p=0.29) patients were readmitted during the primary treatment period. Overall, 42.3% of the readmissions were surgery related, 47.8% were chemotherapy related, and 59.6% were cancer related but not related to surgery or chemotherapy, and each readmission could qualify for more than one reason. Readmitted patients had a higher rate of chronic kidney disease (4.1% vs 1.0%, p=0.038). Post-operative, chemotherapy, and cancer-related readmissions were similar between the two groups. However, the percentage of inpatient treatment days due to unplanned readmission was twice as high for primary cytoreductive surgery at 2.2% vs 1.3% for neoadjuvant chemotherapy (p<0.001). Despite longer readmissions in the primary cytoreductive surgery group, Cox regression analysis demonstrated that readmissions did not affect progression-free survival (HR=1.22, 95% CI 0.98 to 1.51; p=0.08). Primary cytoreductive surgery, higher modified Frailty Index, grade 3 disease, and optimal cytoreduction were associated with longer progression-free survival. In this study, 35% of the women with advanced ovarian cancer had at least one unplanned readmission during the entire treatment time. Patients treated by primary cytoreductive surgery spent more days during readmission than those with neoadjuvant chemotherapy. Readmissions did not affect progression-free survival and may not be valuable as a quality metric.

Secondary validation of an ovarian cancer-specific comorbidity index in a US population

The Ovarian Cancer Comorbidity Index (OCCI) is an age-specific index developed and previously found to be more predictive of overall and cancer-specific survival than the Charlson Comorbidity Index (CCI). The objective was to perform secondary validation of the OCCI in a US population. A cohort of ovarian cancer patients undergoing primary or interval cytoreductive surgery from January 2005 to January 2012 was identified in SEER-Medicare. OCCI scores were calculated with the regression coefficients determined from the original developmental cohort for five comorbidities. Cox regression analyses were used to calculate associations between the OCCI risk groups and 5-year overall survival and 5-year cancer-specific survival in comparison to the CCI. A total of 5052 patients were included. Median age was 74 (range 66-82) years. 47% (n=2375) had stage III and 24% (n=1197) had stage IV disease at diagnosis. 67% had a serous histology subtype (n=3403). All patients were categorized as moderate (48.4%) or high risk (51.6%). The prevalence of the five predictive comorbidities were: coronary artery disease 3.7%, hypertension 67.5%, chronic obstructive pulmonary disease 16.7%, diabetes 21.8%, and dementia 1.2%. Controlling for histology, grade, and age-stratification, worse overall survival was associated with both a higher OCCI (hazard ratio (HR) 1.57; 95% confidence interval (CI) 1.46 to 1.69) and CCI (HR 1.96; 95% CI 1.66 to 2.32). Cancer-specific survival was associated with the OCCI (HR 1.33; 95% CI 1.22 to 1.44) but was not associated with the CCI (HR 1.15; 95% CI 0.93 to 1.43). This internationally developed comorbidity score for ovarian cancer patients is predictive for both overall and cancer-specific survival in a US population. CCI was not predictive for cancer-specific survival. This score may have research applications when utilizing large administrative datasets.

Enhanced recovery for obese patients undergoing gynecologic cancer surgery

To compare post-operative length of stay and complication rates of matched obese and non-obese patients in an enhanced recovery (ERAS) program after open gynecologic cancer surgery. We performed an observational cohort study of patients (n=1225) undergoing open surgery from November 2014 to November 2018 at a tertiary cancer center. Patients undergoing multidisciplinary procedures, non-oncologic surgery, or procedures in addition to abdominal surgery were excluded (n=190). Obese and non-obese patients were matched by date, age, disease status, and surgical complexity. The primary outcome was post-operative length of stay. Secondary outcomes included 30-day peri-operative complications, re-operation, re-admission, opioid use, and program compliance. After matching, 696 patients (348 obese, 348 non-obese) with median age of 57 years (IQR 48-66) were analyzed. Obese patients had a longer median procedure time (218 min vs 192.5 min, p<0.001) and greater median estimated blood loss (300 mL vs 200 mL, p<0.001). Median (IQR) post-operative length of stay was the same for obese and non-obese patients: 3 days (IQR 2-4). Obese and non-obese patients had similar rates of grade III-IV complications (10.9% and 6.6%, respectively, p=0.06), re-operation (2.3% and 1.4%, respectively, p=0.58), and re-admission (11.8% and 8.0%, respectively, p=0.13). Grade I-II complications were more common among obese patients (62.4% vs 48.3%, p<0.001) because they had more wound complications (17.8% vs 4.9%, p<0.001). Obese patients received more opioids both during surgery (morphine equivalent dose 57.25 mg (IQR 35-72.5) vs 50 mg (IQR 25-622.5), p=0.003) and after surgery (morphine equivalent daily dose 45 mg/day (IQR 10-96.2) vs 29.37 mg/day (IQR 7.5-70), p=0.01). Obese and non-obese patients had similar ERAS program compliance (70.1% and 69.8%, respectively, p=0.32). Neither post-operative length of stay nor the rate of serious complications differed significantly despite longer surgeries, greater blood loss, and more opioid use among obese patients. An ERAS program was safe, effective, and feasible for obese patients with suspected gynecologic cancer.

Patient cost sharing during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment in ovarian cancer

More patients with ovarian cancer are being treated with poly(adenosine diphosphate-ribose) polymerase inhibitors because regulatory agencies have granted these drugs new approvals for a variety of treatment indications. However, poly(adenosine diphosphate-ribose) polymerase inhibitors are expensive. When administered as a maintenance therapy, these drugs may be administered for months or years. How much of this cost patients experience as out-of-pocket spending is unknown. This study aimed to estimate the out-of-pocket spending that patients experience during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment and to characterize which healthcare services account for that spending. A retrospective cohort study was performed with a sample of patients with ovarian cancer treated between 2014 and 2017 with olaparib, niraparib, or rucaparib. Patients were identified using MarketScan, a health insurance claims database. All insurance claims during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment were collected. The primary outcome variable was the patients' out-of-pocket spending (copayment, coinsurance, and deductibles) during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment for the medication itself. Other outcomes of interest included out-of-pocket spending for other healthcare services, the types and frequency of other healthcare services used, health plan spending, the estimated proportion of patients' household income used each month for healthcare, and patients' out-of-pocket spending immediately before poly(adenosine diphosphate-ribose) polymerase inhibitor treatment. We identified 503 patients with ovarian cancer with a median age of 55 years (interquartile range, 50-62 years); 83% of those had out-of-pocket spendings during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment. The median treatment duration was 124 days (interquartile range, 66-240 days). The mean out-of-pocket spending for poly(adenosine diphosphate-ribose) polymerase inhibitors was $305 (standard deviation, $2275) per month. On average, this accounted for 44.8% (standard deviation, 34.8%) of the patients' overall monthly out-of-pocket spending. The mean out-of-pocket spending for other healthcare services was $165 (standard deviation, $769) per month. Health plans spent, on average, $12,661 (standard deviation, $15,668) per month for poly(adenosine diphosphate-ribose) polymerase inhibitors and $7108 (standard deviation, $15,254) per month for all other healthcare services. The cost sharing for office visits, laboratory tests, and imaging studies represented the majority of non-poly(adenosine diphosphate-ribose) polymerase inhibitor treatment out-of-pocket spending. The average amount patients paid for all healthcare services per month during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment was $470 (standard deviation, $2407), which was estimated to be 8.7% of the patients' monthly household income. The mean out-of-pocket spending in the 12 months before poly(adenosine diphosphate-ribose) polymerase inhibitor treatment was $3110 (standard deviation, $6987). Patients can face high out-of-pocket costs for poly(adenosine diphosphate-ribose) polymerase inhibitors, although the sum of cost sharing for other healthcare services used during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment is often higher. The spending on healthcare costs consumes a large proportion of these patients' household income. Patients with ovarian cancer experience high out-of-pocket costs for healthcare, both before and during poly(adenosine diphosphate-ribose) polymerase inhibitor treatment.

Bronchocutaneous fistula from metastatic cervical cancer with COVID-19