A data processing pipeline for the AACR project GENIE biopharma collaborative data with the {genieBPC} R package.

MOTIVATION: Data from the American Association for Cancer Research Project Genomics Evidence Neoplasia Information Exchange Biopharma Collaborative (GENIE BPC) represent comprehensive clinical data linked to high-throughput sequencing data, providing a multi-institution, pan-cancer, publicly available data repository. GENIE BPC data provide detailed demographic, clinical, treatment, genomic and outcome data for patients with cancer. These data result in a unique observational database of molecularly characterized tumors with comprehensive clinical annotation that can be used for health outcomes and precision medicine research in oncology. Due to the inherently complex structure of the multiple phenomic and genomic datasets, the use of these data requires a robust process for data integration and preparation in order to build analytic models. RESULTS: We present the {genieBPC} package, a user-friendly data processing pipeline to facilitate the creation of analytic cohorts from the GENIE BPC data that are ready for clinico-genomic modeling and analyses. AVAILABILITY AND IMPLEMENTATION: {genieBPC} is available on CRAN and GitHub.

Pregnancy-associated breast cancer: does timing of presentation affect outcome?

PURPOSE: Pregnancy-associated breast cancer (PABC) comprises breast cancer diagnosed during the gestational period or within 12 months postpartum. While the incidence of PABC appears to be increasing, data regarding prognosis remain limited. METHODS: Here we evaluate clinicopathologic features, treatments, and clinical outcomes among women with stage 0-III PABC diagnosed between 1992 and 2020. Comparisons were made between women who were diagnosed with PABC during gestation and those who were diagnosed within 12 months postpartum. RESULTS: A total of 341 women were identified, with a median age of 36 years (range 25-46). The pregnancy group comprised 119 (35%) women, while 222 (65%) women made up the postpartum group. Clinicopathologic features were similar between groups, with most patients being parous and presenting with stage I and II disease. Treatment delays were uncommon, with a median time from histologic diagnosis to treatment of 4 weeks for both groups. Recurrence-free survival was similar between groups: 67% at 10 years for both. While 10-year overall survival appeared higher in the postpartum group (83% versus 78%, p = 0.02), only the presence of nodal metastases was associated with an increased risk of death (hazard ratio 5.61, 95% CI 2.20-14.3, p < 0.001), whereas timing of diagnosis and receptor profile did not reach statistical significance. CONCLUSION: Clinicopathologic features of women with PABC are similar regardless of timing of diagnosis. While 10-year recurrence-free survival is similar between groups, 10-year overall survival is higher among women diagnosed postpartum; however, timing of diagnosis may not be the driving factor in determining survival outcomes.

Personal Payments from Pharmaceutical Companies to Authors of Oncology Clinical Practice Guidelines.

BACKGROUND: Oncologists who author clinical practice guidelines frequently have financial relationships with the pharmaceutical industry. It is unknown whether participation on clinical practice guideline committees is associated with differences in the amounts of industry money received. MATERIALS AND METHODS: We conducted a nested case-control study from August 2013 to December 2018. We manually abstracted membership records of National Comprehensive Cancer Network (NCCN) Guidelines committees for the 20 most common cancers and linked to Open Payments. The study sample included medical oncologists selected to join an NCCN Guidelines committee ("joiners") during the study period. Joiners were matched 1:2 to medical oncologists who had no participation on NCCN committees (controls) by gender, NCCN institution, and medical school graduation year. We performed difference-in-differences (DiD) estimation to assess whether selection to an NCCN committee was associated with the dollar value of payments received from industry, using generalized estimating equations to address correlation between matched pairs and between repeated observations of the same pair. RESULTS: During the study period, 54 physicians joined an NCCN Guidelines committee. These physicians received more payments than matched controls in the year prior to joining ($11,259 vs. $3,427; p = .02); this difference did not increase in the year after joining (DiD = $731; p = .45). CONCLUSION: Medical oncologists selected to NCCN Guidelines committees had greater financial ties to industry than their peers. The potential influence of industry in oncology clinical practice guidelines may be reduced through the selection of committee members with fewer ties to industry. IMPLICATIONS FOR PRACTICE: Oncologists who author clinical practice guidelines frequently have financial conflicts of interest with the pharmaceutical industry. This creates concern about the potential for industry influence on guidelines. However, it is unknown whether oncologists who author guidelines have greater industry relationships than their peers. This study compared medical oncologists who were newly selected to join a National Comprehensive Cancer Network (NCCN) Guidelines panel with medical oncologists at the same institutions and at similar career stages. At the time they joined, oncologists joining NCCN Guidelines panels had received more than three times the dollar value of industry payments than their peers. The potential for industry influence may be reduced by the selection of less-conflicted panel members.

Association Between Clinical Value and Financial Cost of Cancer Treatments: A Cross-Sectional Analysis.

BACKGROUND: The cost of cancer treatment has increased significantly in recent decades, but it is unclear whether these costs have been associated with commensurate improvement in clinical value. This study aimed to assess the association between the cost of cancer treatment and 4 of the 5 NCCN Evidence Blocks (EB) measures of clinical value: efficacy of regimen/agent, safety of regimen/agent, quality of evidence, and consistency of evidence. METHODS: This is a cross-sectional, observational study. We obtained NCCN EB ratings for all recommended, first-line, and/or maintenance treatments for the 30 most prevalent cancers in the United States and calculated direct pharmacologic treatment costs (drug acquisition, administration fees, guideline-concordant supportive care medications) using Medicare reimbursement rates in January 2019. We used generalized estimating equations to estimate the association between NCCN EB measures and treatment cost with clustering at the level of the treatment indication. RESULTS: A total of 1,386 treatments were included. Among time-unlimited treatments (those administered on an ongoing basis without a predetermined stopping point), monthly cost was positively associated with efficacy ($3,036; 95% CI, $1,782 to $4,289) and quality of evidence ($1,509; 95% CI, $171 to $2,847) but negatively associated with safety (-$1,470; 95% CI, -$2,790 to -$151) and consistency of evidence (-$2,003; 95% CI, -$3,420 to -$586). Among time-limited treatments (those administered for a predetermined interval or number of cycles), no NCCN EB measure was significantly associated with treatment cost. CONCLUSIONS: An association between NCCN EB measures and treatment cost was inconsistent, and the magnitude of the association was small compared with the degree of cost variation among treatments with the same EB scores. The clinical value of cancer treatments does not seem to be a primary determinant of treatment cost.

The Accuracy and Usefulness of the National Comprehensive Cancer Network Evidence Blocks Affordability Rating.

BACKGROUND: The National Comprehensive Cancer Network (NCCN) Guidelines' Evidence Blocks has the broadest scope of the several oncology value assessment frameworks. The Evidence Blocks includes the Affordability criterion, which reflects the financial cost of each treatment on a 1-5 scale. The accuracy of Affordability is unknown. METHODS: We calculated Medicare costs for all first-line and maintenance treatments for the 30 cancers with the highest incidence in the USA that had published NCCN Evidence Blocks as of 31 December 2018. We assessed the accuracy and consistency of Affordability across different treatments and cancer types. Among different treatments for the same indication, we determined the frequency with which the Affordability assessment was consistent with calculated treatment costs. RESULTS: There were a total of 1386 treatments in our sample. Lower Affordability scores were associated with higher costs. There was significant variation in cost at each level of Affordability; for treatments with Affordability = 1 (very expensive), costs ranged from $US4551 to $US43,794 per month for treatments administered over an undefined time period and from $US2865 to $US500,982 per course of therapy for treatments administered over a defined time period. Among treatments for the same indication, Affordability was discrepant with calculated treatment costs in 7.9% of pairwise comparisons, identifying the higher-cost treatment as being more affordable. Discrepancies were reduced when we reassigned Affordability scores based on calculated treatment costs. CONCLUSIONS: Evidence Blocks Affordability generally correlated with treatment costs but contained discrepancies, which may limit its usefulness to clinicians in comparing costs. This study suggests that the Affordability score may be improved by indexing more directly to specified dollar value thresholds.