In this retrospective study, demographic, clinical, and treatment data were primarily abstracted from the McKesson Specialty Health/US Oncology Network (MSH/USON) iKnowMed (iKM) electronic health record (EHR) database. iKM is an oncology-specific, integrated, web-based system that captures outpatient practice encounter history for patients treated across 19 US states and at over 400 sites of care. Overall, the iKM EHR database captures data from approximately 10% of newly-diagnosed cancer patients in the USA (~750,000) annually and includes 1000 affiliated physicians. MSH/USON’s Claims Data Warehouse (CDW), which records all health insurance claims-related information, was used to obtain payer data. Documented vital status (death) in iKM was supplemented with data from the Social Security Death Index (SSDI). In addition, chart reviews were conducted by clinical staff to supplement and validate data captured by programmatic queries for prespecified endpoints from the iKM EHR.
Patient Selection Criteria
Patients with HER2-positive mBC who initiated treatment with pertuzumab within the study period of 1 June 2012, and 30 June 2014 were eligible for inclusion in the study. All patients were followed through 31 December 2014 (end of study follow-up). The primary analysis population met the following criteria: (1) female patients with HER2-positive mBC; (2) had at least two visits within USON during the study period; (3) received care at a practice utilizing the full EHR capabilities of iKM; (4) ≥18 years old at diagnosis; and (5) received pertuzumab within 60 days of mBC diagnosis during the study period. Patients were excluded if they were diagnosed with or treated for another primary cancer during the study period, or were concurrently enrolled in a clinical trial.
The date of initiation of pertuzumab was defined as the index date. PFS was defined as the time from the index date until the first disease progression, or death from any cause, whichever occurred earlier. Evidence of progression was captured from chart review. Patients without a progression or death event were censored at the last pertuzumab administration date within the study period, last visit date, or the end of study-follow up, whichever occurred first.
Adverse events were captured from the index date until 30 days after the pertuzumab discontinuation date, or the end of study follow-up, whichever occurred first. Pertuzumab treatment duration was defined as the time from the index date to the last pertuzumab administration date, last visit date, death date, or the end of study-follow up, whichever occurred first.
The Eastern Cooperative Oncology Group (ECOG) performance status at index was recorded for each patient. Comorbidities noted within 6 months prior to the index date were included, and assigned a weight based on the Charlson Comorbidity Index (CCI). Disease type at diagnosis was classified as non-visceral if the metastatic sites were breast, bone/bone marrow, lymph node, skin, or chest wall, or visceral if the metastatic sites were adrenal gland, ascites, brain, cervix, esophagus, gall bladder/bile duct, head and neck localized lesion, kidney, liver, lung, mediastinum, ovary, pancreas, pelvis (non-bone), pericardial effusion, rectum, trachea/bronchi, uterus, or other. Patients with both visceral and non-visceral metastatic sites were considered to be a part of the visceral subset. Among patients with recurrent disease, disease-free interval (DFI) was calculated as the duration from date of primary breast cancer surgery to mBC diagnosis.
All statistical analyses were conducted on the primary analysis population. Descriptive statistics were utilized to summarize demographics, clinical characteristics, treatment patterns, and AEs. Categorical variables were reported as frequency and percentage, and continuous variables were reported as mean, standard deviation, median, and range. Since docetaxel was the taxane used in the CLEOPATRA pivotal trial [13, 14], and as trastuzumab + pertuzumab + docetaxel is also the FDA-approved indicated regimen in the mBC setting , a sensitivity analysis was performed on a subset of the primary analysis population who initiated treatment with trastuzumab + pertuzumab + docetaxel. Kaplan–Meier (KM) survival analyses were conducted to estimate median PFS, and PFS rates with their corresponding 95% confidence intervals (CIs) at 6, 9, 12, and 18 months, for the primary and sensitivity analysis populations. All statistical analyses were performed using SAS® 9.2 (SAS Institute, Cary, NC, USA).