TO THE EDITOR: Chae et al. (1) prospectively investigated the ability of pretreatment 18F-fluoroestradiol PET/CT to predict the pathologic response to neoadjuvant therapy in postmenopausal women with estrogen receptor–positive breast cancer. Of 25 evaluated patients, 12 received neoadjuvant chemotherapy and 13 neoadjuvant endocrine therapy. In the former group, the 2 patients with 18F-fluoroestradiol–negative tumors and none of the 10 patients with 18F-fluoroestradiol–avid tumors achieved a pathologic complete response (P = 0.02). In the latter group, all 13 patients had 18F-fluoroestradiol–avid uptake, but none achieved a pathologic complete response. No difference in pretreatment SUVmax between responders and nonresponders was observed in either group. However, using the Miller–Payne grading system to define response, 5 of 7 neoadjuvant chemotherapy patients with a baseline SUVmax of less than 7.3 achieved a pathologic response, whereas none of the 5 neoadjuvant endocrine therapy patients with an SUVmax of less than 7.3 were responders (P = 0.03). In agreement with another study (2), these results suggest that patients with low tumor 18F-fluoroestradiol uptake at baseline are more likely to be treated with neoadjuvant chemotherapy than with neoadjuvant endocrine therapy. In patients with a high baseline tumor SUVmax, Chae et al. observed no difference in pathologic response, whatever the treatment group. For these tumors with high 18F-fluoroestradiol uptake, a second PET examination could potentially be helpful to measure the change in SUV under treatment, in the same way as is sometimes done with 18F-FDG imaging (3). This could increase the predictive value of 18F-fluoroestradiol imaging. In the metastatic setting, among 16 patients treated with fulvestrant, baseline 18F-fluoroestradiol PET was unable to predict the response (4). When a second examination was performed a few weeks after the start of treatment, the change in tumor 18F-fluoroestradiol uptake was significantly larger in patients having clinical benefit from fulvestrant than in patients with progressive disease (P = 0.025) (4). Another research possibility would be the use of 18F-FDG imaging in addition to 18F-fluoroestradiol PET. In estrogen receptor–positive breast cancer, recent studies suggested that 18F-FDG uptake measured at a single time point before neoadjuvant chemotherapy (5) or before initial surgery (6) was associated with patient survival. A pilot study evaluated the value of 18F-fluoroestradiol and 18F-FDG imaging together in predicting the response of various breast cancer phenotypes to neoadjuvant chemotherapy. The ratio of 18F-fluoroestradiol SUV to 18F-FDG SUV showed great value in predicting the response (P = 0.002) (2). However, the small number of patients in this study (n = 18) was a limitation. Moreover, luminal tumors were mixed with estrogen receptor–negative breast cancer. In the metastatic setting, the recent study from Kurland et al. showed that information from baseline 18F-fluoroestradiol and 18F-FDG imaging can be used together to separate patients into 3 groups with different prognoses (7).
In conclusion, although the study from Chae et al. suggests that baseline 18F-fluoroestradiol PET could be of interest to predict the response to neoadjuvant therapy in estrogen receptor–positive breast cancer patients, the predictive value seems to have some limitations. Performing a second 18F-fluoroestradiol PET examination during treatment or complementing the 18F-fluoroestradiol examination with a 18F-FDG PET examination could potentially improve the predictive power and deserves to be evaluated prospectively in a large study.
Footnotes
Published online Nov. 10, 2016.
- © 2017 by the Society of Nuclear Medicine and Molecular Imaging.