Pre-Treatment Staging Imaging in Rectal Cancer: Results From the Quality Oncology Practice Initiative

For patients with nonmetastatic rectal cancer, the National Comprehensive Cancer Network guidelines recommend initial staging using pelvic magnetic resonance imaging or endorectal ultrasound to determine the stage of the disease before initial therapy or surgery. This imaging workup helps determine the T and N staging, which is essential to determine optimal treatment for a patient. The current study examined practice concordance with this guideline using a quality measure in ASCO's Quality Oncology Practice Initiative (QOPI) that specifically addressed staging workup for patients with rectal cancer.

From Fall 2016 through Fall 2019, 103 QOPI-participating practices reported performance on QOPI measure Colorectal 78, which addresses staging workup for patients with rectal cancer. The percentage and 95% CI of patients who received guideline-concordant imaging were calculated for each of the seven assessment time points. Difference of concordance rates between subsequent time points and the initial time point was assessed using logistic regression with random-effects models. In addition, 69 practices that submitted data in 2016 and 2017 were surveyed to gain insight on potential reasons for nonconcordance and the results were described.

At each time point, a total of 20-33 practices reported data across 1,158 unique patients. Adherence appeared to increase over time, with 38% of patients receiving guideline-recommended staging in Fall 2016, to 56% in Fall 2019. The practice survey revealed that nonconcordance was mostly because of lack of care coordination between oncology and surgery disciplines (n = 16 practices of 28, 57.1%) and a lack of awareness of appropriate staging scans (n = 8 practices, 28.6%).

As one half to one third of patients still do not receive appropriate imaging, our findings highlight the need for concerted quality improvement efforts that involve the multidisciplinary team to close this gap.