Our population-based study examined the patterns of first treatment for patients with TN and HER2 breast cancer before patterns of care were disrupted by the COVID19 pandemic. Among patients with TN and HER2 breast cancer, approximately 1 out of 5 underwent NAC rather than surgery as the first treatment for breast cancer. In patients with tumors ≥ 2cm, only 39% had a medical oncology consultation prior to first treatment with 83% of these patients ultimately undergoing NAC. Interestingly, these numbers suggest that when referred to medical oncology, a high proportion of patients do undergo NAC as first treatment. Conversely, 95% of patients with TN and HER2 breast cancer who were first assessed by a surgeon and underwent upfront surgery, did so without a medical oncology consultation. These results suggest early medical oncology referral is an essential component of multidisciplinary care for patients with TN and HER2 breast cancers, when considering NAC as the evidence-based standard of care for these patients.
The use of NAC in the management of breast cancer has been examined in previous retrospective work using population-based and multi-institutional data [14–19]. These studies demonstrate variability in the rates of NAC based on geographic jurisdiction, breast cancer type and tumor characteristics. In the United States, 17% of all patients and 23% of patients with ER negative cancers underwent NAC [16]. NAC was utilized more frequently in younger patients with node positive or more advanced stage cancers and at academic centres, with usage reported up to 79% in stage IIIB cancers in the National Cancer Database during 2006–2012 [17]. Outside North America, the reported use of NAC ranged from 10–58% during study periods between 2008 and 2017 [18, 19, 36]. Despite the range of NAC use reported in the literature, there are no reported targets or benchmarks for NAC use in patients with TN and HER2 breast cancers. Approximately 60% of our cohort had tumors ≥ 2cm and 40% of our cohort had node positive disease. Given current recommendations, one would expect NAC rates to be higher than reported based on these characteristics. Given the lack of benchmarks, it is challenging to determine if reported rates in our study and the existing literature are lower or higher than to be expected.
Within a similar cohort to ours, Brezden-Masley et al identified NAC rates of 20.1% and 18.7% between 2012 and 2016 for patients with HER2 and TN breast cancers, respectively [14, 15]. In 2015, the CREATE X trial was presented at the San Antonio breast cancer conference. This trial, which was then published in 2017, demonstrated a survival advantage for adjuvant capecitabine in TN breast cancer patients who underwent NAC without pCR on final pathology [8]. The KATHERINE trial, which identified an invasive disease-free survival advantage for adjuvant TDM-1 in HER2 patients after NAC without pCR, was presented and published in 2019 [9]. These studies added support for NAC in patients with HER2 and TN breast cancer to allow for both prognostication, and response-driven therapies, based on post-NAC pathology. Given the dissemination timeline of these practice changing outcomes, we expected that the rates of NAC would be increased during our study period which extended to March 31 2019. While the rates of NAC use were slightly higher for each receptor group than previously reported (HER2 23.8%, TN 21.7%), it was interesting to find that a large proportion of patients were not even assessed by medical oncologists for consideration of NAC. This finding may be a key element that requires further evaluation. Any identified gaps and barriers in this area (i.e. referral to medical oncology), may help develop interventions to facilitate improved NAC consideration for these patients. To our knowledge, this study is the first in the published literature to focus specifically on NAC uptake in TN and HER2 positive cancers and highlight that evaluation by medical oncology practitioners may be an important driver of this measure.
While medical oncology consultation is essential to the administration of NAC, surgeons are typically the first point of contact for patients after their breast cancer diagnosis. This point was true in our study as approximately 98% of patients underwent their first consultation after diagnosis by a surgeon. However, it is unclear from our data if the low rates of medical oncology consultations for patients undergoing surgery first was because surgeons did not refer to medical oncologists or if the referrals were declined by medical oncologists. Importantly, patients who underwent their first consultation at a RCC were more likely to receive a medical oncology consultation prior to initiating therapy. This points towards the role of access to medical oncology services as a potential factor in the referral for, and receipt of NAC. While this analysis was beyond the scope of the current study, an understanding of how the distribution, availability, and accessibility of both medical oncology and chemotherapy services influence the use of NAC, will be needed to design strategies and clinical pathways to increase access to NAC for more patients. Indeed, geographic variation in care, including medical oncology consultation, has been reported for other cancers, even within universal health systems [19, 37–39].
This study has limitations. First, the data used were from health care administrative datasets and not collected specifically to answer the research question. Billing and administrative data may not be representative of the exact procedure or treatment rendered; in addition, missing data for which records are not available cannot be accounted for. Another limitation of our data is the lack of both clinical staging and surgical pathologic data for patients undergoing NAC versus surgery. For example, some patients staging classification was based on clinical staging while others were based on their surgical pathology. Typically, clinical staging was used for patients undergoing NAC first and pathologic staging was used for patients undergoing surgery first. In addition, we were unable to attribute a date to the receptor data. Therefore, it is not possible from our administrative data to elicit if the receptor status was known or available at the time of first consultation or prior to surgery. Within the province, the reporting of receptors on core biopsy is not mandated and therefore, some centres perform receptor testing routinely on core biopsies, while others rely on the physicians to specifically request the receptor testing. This discrepancy means that some physicians may be making clinical decisions with incomplete information. Of note, we only included patients that underwent surgery after NAC, introducing a potential selection bias as patients who progressed on treatment, developed metastatic disease, opted not to pursue surgery, or left the province will not be captured. Lastly, patient level information regarding patient wishes or preference regarding treatment approach, and performance status, are also not captured in the retrospective administrative data.