Use of 64Cu-DOTA-Trastuzumab PET to Predict Response and Outcome of Patients Receiving Trastuzumab Emtansine (T-DM1) for Metastatic Breast Cancer: A Pilot Study

Trastuzumab emtansine (T-DM1) is an antibody drug conjugate used to treat HER2-positive breast cancer. We hypothesized that functional imaging with 64Cu-DOTA-trastuzumab PET/CT would predict patient response to T-DM1 therapy. Methods: Ten women with biopsy-proven HER2-positive metastatic breast cancer who were to be treated with T-DM1 underwent pretreatment 18F-FDG PET/CT. 64Cu-DOTA-trastuzumab PET/CT was performed 1 and 2 days post-injection. Treatment outcome was assessed by (a) response to T-DM1 (3.6 mg/kg every 3 wk) as evaluated by follow-up 18F-FDG PET/CT using PERCIST criteria and (b) time to treatment failure (TTF). Tumors measurable for response were evaluated for 64Cu-DOTA-trastuzumab uptake in terms of maximum voxel standardized uptake value (SUVmax, units g/mL). Results: Response was clearly related to tumor uptake of 64Cu-DOTA-trastuzumab. Comparing group means, responsive patients (n = 5) had higher day 1 minimum SUVmax (5.6 vs 2.8, P < 0.02), day 2 average SUVmax (8.5 vs 5.4, P < 0.05) and day 2 minimum SUVmax (8.1 vs 3.2, P < 0.01) than non-responsive patients (n = 5 and 4 for days 1 and 2, respectively; inadvertently, 1 patient's day 2 scan included no tumors measurable for response). Tumor-level response suggested threshold dependence on SUVmax Optimal SUVmax cutpoints correctly separated 13/18 responsive and 11/13 non-responsive tumors on day 1, as well as 14/15 responsive and 9/10 non-responsive tumors on day 2. Although not statistically significant in a hierarchical model of tumors-within-patient in this 10 patient cohort, the SUVmax response threshold estimates for individual tumors accurately separated patients with respect to response and TTF. On day 1, 4/5 responsive and 0/5 non-responsive patients had minimum SUVmax above threshold (P < 0.05). On day 2, average SUVmax was above threshold for 5/5 responsive and 1/4 non-responsive patients (P < 0.05), while 5/5 responsive and 0/4 non-responsive patients had minimum SUVmax > threshold (P < 0.01). Patients with day 2 minimum SUVmax above versus below threshold had median TTF = 28 months versus 2 months (P < 0.02).