Combining routine histologic information, including tumor size, grade, and cellular proliferation marker Ki67, failed to accurately reflect the recurrence score produced by the commercial diagnostic test, Oncotype Dx, according to researchers.
Although each variable individually was significantly correlated with Oncotype Dx, the combination of variables had a percentage of variance (R2) of 0.35 on the recurrence score and did not accurately predict the result generated by the 21-gene reverse-transcriptase polymerase chain reaction (RT-PCR) diagnostic test.
The single-institution study included 252 patients, (248 females and 4 males, median age 56 years) with early-stage, estrogen receptor–positive breast cancer. All patients had Oncotype Dx testing performed between 2007 and 2014. The median tumor size was 2.1 cm, and 49% were grade 2, 28% grade 3, and 23% grade 1.
Investigators examined baseline patient demographic data, such as age, race, and sex, and routine pathologic features, such as histologic grade, Ki-67, tumor size, and histologic type. The combination of three variables that individually were highly correlated to the Oncotype Dx score did not produce values that could substitute for the recurrence score.
Furthermore, the prediction capability of the three-variable combination decreased for tumors with high-recurrence scores.
“Therefore, we conclude the RS [recurrence score] can provide additional valuable information and should not be replaced by analysis of routine histologic variables alone,” wrote Dr. Kate Lathrop of the Cancer Therapy and Research Center at the University of Texas Health Science Center in San Antonio, and her colleagues.