38 Clinicopathological Risk Factors, Poorly Stratified Baseline Risk, and RT Benefit Compared to DCISionRT in Patients With Ductal Carcinoma in Situ

, , , , , , , ,
Miami Breast Cancer Conference® Abstracts Supplement, 39th Annual Miami Breast Cancer Conference® - Abstracts, Volume 36, Issue suppl 3
Pages: 27

Objective

Over- and undertreatment of ductal carcinoma in situ (DCIS) has been a persistent clinical problem with traditional clinicopathologic features unable to stratify patients appropriately. DCISionRT reports a decision score (DS) validated in multiple cohorts, including the randomized SweDCIS Trial, which is prognostic for 10-year ipsilateral breast recurrence (IBR) risk after breast-conserving surgery (BCS) and predictive for radiotherapy (RT) benefit. Here, stratification of IBR risk and RT benefit by clinicopathology was assessed with DCISionRT.


Methods

DCISionRT and its integrated residual risk subtype (RRt) was evaluated in 493 patients from 3 cohorts at a Clinical Laboratory Improvement Amendments (CLIA) lab (PreludeDx, Laguna Hills, CA) who were treated with BCS or BCS plus RT. Clinicopathology features were analyzed in a multivariable analysis with DCISionRT and its integrated RRt-classified risk groups for 10-year IBR.


Results

The biosignature classified patients into low (DS ≤ 2.8, n = 173), elevated (DS > 2.8 without RRt, n = 209), and residual (DS > 2.8 with RRt, n = 111) risk groups. Patients had increased 10-year rates of IBR risk in the elevated (21.4%, P <.001) and residual (43.4%, P <.001) risk groups without RT, and benefited from RT (elevated: HR, 0.15, P = .002; residual: HR, 0.27, P = .004) versus the low-risk group (5.5%; HR, 1.28; P = .7). The residual versus elevated risk group had an increased 10-year IBR risk rate after RT (20.5% vs 3.2%, P = .008). Clinicopathologic features (age, grade, size, palpability, and necrosis) were not associated with the 10-year IBR risk rate in multivariable analysis including DCISionRT and treatment. The distribution of clinicopathologic features varied between biosignature risk groups; the residual risk group had a higher proportion of patients with nuclear grade 3 (71% vs 31%, P < .001), necrosis (89% vs 56%, P < .001), and size greater than 1 cm (54% vs 35%, P < .001) when compared with low- and elevated- risk groups, but no significant change was noted in the distribution between DCISionRT risk groups for age.


Conclusions

DCISionRT classified patients into 3 groups with distinct risk and RT benefit profiles, where the residual risk group had the highest 10-year IBR risk rate without RT and significantly elevated IBR risk after RT. The distribution of clinicopathological features varied between biosignature risk groups but were not significantly associated with the 10-year IBR risk rate, accounting for DCISionRT risk groups and treatment.

Author Affiliations:

Pat W. Whitworth,1 Rachel Rabinovitch,2 Frank A. Vicini,3 Chirag Shah,4 Fredrik Wärnberg,5 G. Bruce Mann,6 Steven C. Shivers,7 Karuna Mittal,7 Troy Bremer,7

1Nashville Breast Center, Nashville, TN
2University of Colorado, Colorado Springs, CO 3GenesisCare, Farmington Hills, MI
4Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH

5Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
6University of Melbourne, Melbourne, Australia

7PreludeDx, Laguna Hills, CA