Clin Nucl Med. 2017 Mar;42(3):169-175.

Evaluation of the Response to Breast Cancer Neoadjuvant Chemotherapy Using 18F-FDG Positron Emission Mammography Compared With Whole-Body 18F-FDG PET: A Prospective Observational Study.

Noritake M1, Narui K2, Kaneta T1, Sugae S3, Sakamaki K4, Inoue T1, Ishikawa T5.

1 Department of Radiology, Graduate School of Medicine, Yokohama City University, Yokohama, Japan.

2 Department of Breast and Thyroid Surgery, Yokohama City University Medical Center, Yokohama, Japan.

3 Departments of Gastroenterological Surgery, Graduate School of Medicine, Yokohama City University, Yokohama, Japan.

4 Departments of Biostatics and Epidemiology, Graduate School of Medicine, Yokohama City University, Yokohama, Japan.

5 Department of Breast Surgery, Tokyo Medical University, Tokyo, Japan.

 

Abstract

PURPOSE: The aim of this study was to assess therapeutic response to breast cancer neoadjuvant chemotherapy (NAC) by F-FDG positron emission mammography (PEM) compared with that to whole-body F-FDG PET (WBPET).

METHODS: Twenty patients underwent WBPET and PEM 3 times: the first time was before NAC, the second time was after 2 courses of NAC, and the third time was after all courses of NAC. A pathological complete response (pCR) was defined as no evidence of residual invasive cancer with or without ductal carcinoma in situ. The relationships between each modality’s SUVmax and pathological response were evaluated.

RESULTS: Nine patients achieved a pCR, whereas the other 11 patients had a non-pCR. The SUVmax of WBPET after 2 courses of NAC was significantly lower in the pCR group than in the non-pCR group (1.4 ± 0.4 vs 2.7 ± 2.1, P = 0.0334). There were no significant differences in the SUVmax of PEM (ie, PEM uptake value [PUV]) between the groups. The SUVmax of WBPET (area under the ROC curve [AUC] = 0.761) was superior to the PUVmax (AUC, 0.648) for predicting non-pCR at the interim time point. After all courses of chemotherapy, there were no significant differences between the groups in the SUVmax of WBPET; however, PUVmax was significantly lower in the pCR group than in the non-pCR group (1.0 ± 0.2 vs 2.5 ± 2.7, P = 0.0351). After NAC, the PUVmax (AUC, 0.796) was superior to the SUVmax of WBPET (AUC, 0.671).

CONCLUSIONS: There proved to be no apparent superiority of PEM in predicting pCR at the interim time point. Positron emission mammography had greater diagnostic capability for detecting residual cancer after all courses of NAC.

 

Supplement

Fluorine-18 fluorodeoxyglucose (FDG) positron emission mammography (PEM) is a useful diagnostic modality for breast cancer. The aim of this study was to assess therapeutic response to breast cancer neoadjuvant chemotherapy (NAC) by PEM, compared with that to whole body 18F-FDG positron emission tomography (WBPET).

Previous studies have primarily defined pathologic complete response (pCR) in two ways when assessing metabolic the response to neoadjuvant NAC by using WBPET [1-4]. In some studies, pCR was defined as no evidence of residual invasive tumor, except ductal carcinoma in situ (DCIS) [1,2], and in other studies pCR was defined as no evidence of residual malignancy at all, including no DCIS [3,4]. We used the former definition for the present study. However, we also analyzed our data by the latter definition of pCR (i.e., no evidence of residual malignancy at all). We named this status as pCR2, and we also compared the SUVmax and PUVmax of pCR2 to those of non-pCR2. The outomes are described in Figure1. and Figure2.

Three WBPET and PEM scans were performed for each patient: the first scan at baseline, the second scan after the second chemotherapy cycle (i.e., immediately before the third cycle), and the third scan after all courses of NAC but before surgical treatment.

The standardized uptake value (SUV) of whole body PET was calculated, based on the measured activity, the decay-corrected injected dose, and the patient’s body weight as follows:

SUV = tissue radioactivity concentration (kBq/mL)/[injection dose (MBq)/patient’s weight (kg)].

For PEM, the maximum uptake was recorded as the maximum PEM uptake value (PUVmax). The PUVmax was calculated as follows:

PUV = tissue radioactivity concentration (kBq/mL)/[injection dose (MBq)/patient’s weight (kg)].

The PEM does not have decay-correction technology. Therefore, manufacturers recommend the lesion to background (LTB) ratio as the objective value. The LTB value, the ratio of a lesion’s PUVmax to the mean background value, was then obtained. The LTB ratios were calculated as follows:

LTB = PUVmax of the lesion/PUVmean of the background.

The outcomes are similar to that of our present paper : SUVmax was the best indicator at the interim and PUVmax was the best indicator after all course of NAC. However, the diagnostic accuracy of an examination was higher. The AUC of SUVmax and PUVmax was 0.987 and 0.917. After all courses of NAC, the PUVmax was highly accurate: the AUC was 0.900. The optimal cutoff value was 1.1 for interim SUVmax; sensitivity, 100%; specificity, 60%; positive predictive value, 88%; negative predictive, value 100%; and accuracy, 90%.

We could set a cutoff value for the interim PUVmax and SUVmax. The optimal cutoff value was 1.5. The sensitivity was 87%; specificity, 75%; positive predictive value, 93%; negative predictive value, 60%; and accuracy, 80%.

After all courses of NAC, we set cutoff value of 1.1 for PUVmax (sensitivity, 80%; specificity, 100%; positive predictive value, 100%; negative predictive value, 58%; accuracy, 80%).

If we defined pCR as no evidence of residual malignancy at all, including no DCIS, the PUVmax and LTB were also useful indicators at the interim and after NAC.

 

 

Figure1. Change of SUVmax and PUVmax in pCR2 and non-pCR2 groups at baseline, interim and afterNAC.

Baseline: Before NAC

Interim: After 2 courses of NAC

After NAC: After all courses of NAC

 

 

Figure2. The ROC curves of the interim SUVmax of WBPET, PUVmax, and LTB ratios of PEM for predicting and detecting non-pCR2 at each time point.

Interim: After 2 courses of NAC

After NAC: After all courses of NAC

 

References

1. Keam B, Im SA, Koh Y, et al. Early metabolic response using FDG PET/CT and molecular phenotypes of breast cancer treated with neoadjuvant chemotherapy. BMC Cancer. 2011;11:452.
2. Hatt M, Groheux D, Martineau A, et al. Comparison between 18F-FDG PET image-derived indices for early prediction of response to neoadjuvant chemotherapy in breast cancer. J Nucl Med. 2013;54:341–349.
3. Andrade WP, Lima EN, Osorio CA, et al. Can FDG-PET/CT predict early response to neoadjuvant chemotherapy in breast cancer? Eur J Surg Oncol. 2013;39:1358–1363.
4. Pahk K, Kim S, Choe JG. Early prediction of pathological complete response in luminal B type neoadjuvant chemotherapy-treated breast cancer patients: comparison between interim 18F-FDG PET/CT and MRI. Nucl Med Commun. 2015;36:887–891.