Breast cancer prediction tools were associated with tumor prognosticators such that women with calculated high risk were significantly more likely to be diagnosed with low-grade, estrogen receptor (ER)–positive cancers, according to researchers.
The Tyrer-Cuzick model incorporates hormonal, lifestyle, and reproductive risk factors, as well as family history, into breast cancer risk estimates and was associated with ER status (ER-negative odds ratio, 0.80), grade (grade 3: OR, 0.79), and lymph node involvement (lymph node positive: OR, 0.77), but not tumor size. A polygenic risk score based on 77 single-nucleotide polymorphisms (SNP) variants was associated with ER status (ER-negative: OR, 0.80), tumor size (greater than 40 mm: OR, 0.86), and grade (grade 3: OR, 0.86). Tyrer-Cuzick model associations were observed in women younger than age 50, but not in older women. The polygenic risk score associations held for all age groups.
“Our results support the hypothesis that breast cancer subtypes have different etiologies and highlight the need to identify risk factors separately for distinct breast cancer subtypes and ages of onset. Better knowledge of subtype-specific risk factors and understanding of disease etiology may be vital for the success of primary prevention and screening programs aimed at lowering mortality,” wrote Johanna Holm of the Karolinska Institute, Stockholm, and colleagues (Journ Clin Onc. 2015 Dec 2. doi: 10.1200/JCO.2015.63.0624).
The study evaluated 5,500 female breast cancer patients younger than 80 years diagnosed between 2001 and 2008 in Sweden. In total, 5,232 participants had information on the Tyrer-Cuzick score, 4,927 had information on the polygenic risk score, and 3,488 had a prediagnostic mammographic density measurement.
The researchers assessed three breast cancer prediction tools, the Tyrer-Cuzick–predicted 10-year breast cancer risk score, the polygenic risk score, and mammographic density, to determine if risk estimates differed according to tumor prognosticators and metastasis. Unlike the Tyrer-Cuzick score and polygenic risk score, mammographic densities did not vary according to tumor prognosticators and therefore seem to indicate general breast cancer risk, according to investigators.
Both the Tyrer-Cuzick and polygenic risk scores were associated with favorable tumor prognosticators, and women with high scores in both may be even more likely to have favorable disease outcomes, according to researchers. In support of this finding, the survival analysis showed that women above the median in both risk calculators had decreased risk of distant metastasis.