2366. Effect of FDG-Avid Nodal Disease at Initial Presentation on Survival in Patients with Inflammatory Breast Cancer
Authors * Denotes Presenting Author
  1. Andrea Tenreiro Carneiro; University of Texas MD Anderson Exam
  2. Jia Sun; University of Texas MD Anderson Exam
  3. Megumi Kai; University of Texas MD Anderson Exam
  4. Miral Patel; University of Texas MD Anderson Exam
  5. Sadia Saleem; University of Texas MD Anderson Exam
  6. Gary Whitman; University of Texas MD Anderson Exam
  7. Huong Le-Petross *; University of Texas MD Anderson Exam
The objective of this study is to investigate the predictive effect of FDG-avid nodal disease on overall survival (OS) at initial presentation in patients with inflammatory breast cancer (IBC), compared to nodal ultrasound (US).

Materials and Methods:
Patients with IBC were identified from an institutional review board approved IBC registry at a single institution between October 2006 and December 2019. Informed consent was obtained from all patients prior to entry into the IBC registry. Retrospective chart and imaging reviews of US and PET/CT were performed by two breast radiologists with over 20 years of experience and 1 breast imaging fellow. A waiver of informed consent was approved. Cox regression was used to assess the association of imaging features with OS. Survival curves were estimated using the Kaplan-Meier method. PET/CT and US findings were compared using McNemar’s exact test.

All 262 patients with IBC were identified in the IBC registry database. Mean age at diagnosis was 51 years [range 20-78]. A total of 209 (79.8%) were Caucasian, 19 (7.3%) Black or African American, 10 (3.8%) Asian, 1 (0.4%) American Indian/Alaska Native, and 23 patients identified themselves as other. Histology was invasive ductal carcinoma in 217/262 (82.8%), invasive lobular carcinoma in 11/262 (4.2%), invasive mixed carcinoma in 13/262 (5.0%), invasive non-specified in 21/262 (8%). Mean OS was 3.7 years (CI 3.16-5.11). Axillary adenopathy on US was detected in 160/262 (61%) patients at level I, 94 (35.8%) at level II, and 59 (22.5%) at level III. Ipsilateral axillary adenopathy on PET/CT was detected in 157 (59.9%) at level I, 96 (36.6%) at level II, 60 (22.9%) at level III; no statistical significant difference between US and PET/CT (p>0.05). On PET/CT, mean nodal SUV of the index node was 10 (range 1.2 – 32.7). Ipsilateral internal mammary adenopathy (IMN) was detected on US in 27/262 (10.3%) patients and PET/CT in 44 (16.8%). Contralateral IMN was detected on PET/CT in 8 (3.1%); contralateral IMN US was not routinely performed. Ipsilateral supraclavicular (SC) adenopathy was detected on US in 67/262 (25.6%) and on PET/CT in 65 (24.8%). Contralateral SC adenopathy was detected in 13 (5.0%); no US of contralateral SC performed. On PET/CT, patients with contralateral SC and axillary adenopathy had worse prognosis (median OS = 2.96 years), compared to those without contralateral adenopathy (median OS = 12.53 years), hazard ratio of 3.1 (CI 1.606-5.988), p = 0.003. Median OS was 1.51 years for those with contralateral SC adenopathy, compared to 3.99 years for those without SC adenopathy (p = 0.00038).

Current pre-treatment imaging guidelines recommend mammography and ultrasound first, with PET/CT as optional. Our data suggests that PET/CT should be performed on all IBC patients at initial presentation or earlier in the staging work-up, with US being a bilateral exam as well as to facilitate biopsy of contralateral SC and axillary adenopathy, which significantly reduces overall survival.