¹⁸F-Alfatide II PET/CT for Identification of Breast Cancer: A Preliminary Clinical Study

Abstract

¹⁸F-alfatide II has been proven to have excellent clinical translational potential. In this study, we investigated ¹⁸F-alfatide II for identifying breast cancer and compared the performances between ¹⁸F-alfatide II and ¹⁸F-FDG. Methods: Forty-four female patients with suspected primary breast cancer were recruited. PET/CT images using ¹⁸F-alfatide II and ¹⁸F-FDG were acquired within 7 d. Tracer uptake in breast lesions was evaluated by visual analysis, and semiquantitative analysis with SUV_max and SUV_mean. Results: Forty-two breast cancer lesions and 11 benign breast lesions were confirmed by histopathology in 44 patients. Both ¹⁸F-alfatide II and ¹⁸F-FDG had higher uptake in breast cancer lesions than in benign breast lesions (P < 0.05 for ¹⁸F-alfatide II, P < 0.05 for ¹⁸F-FDG). The area under the curve of ¹⁸F-alfatide II was slightly less than that of ¹⁸F-FDG. Both ¹⁸F-alfatide II and ¹⁸F-FDG had high sensitivity (88.1% vs. 90.5%), high positive predictive value (88.1% vs. 88.4%), moderate specificity (54.5% vs. 54.5%), and moderate negative predictive value (54.5% vs. 60.0%) for differentiating breast cancer from benign breast lesions. By combining ¹⁸F-alfatide II and ¹⁸F-FDG, the sensitivity and negative predictive value significantly increased to 97.6% and 85.7%, respectively, with positive predictive value slightly increased to 89.1% and no change to the specificity (54.5%). The uptake of ¹⁸F-alfatide II (SUV_max: 3.77 ± 1.78) was significantly lower than that of ¹⁸F-FDG (SUV_max: 7.37 ± 4.48) in breast cancer lesions (P < 0.05). ¹⁸F-alfatide II uptake in triple-negative subtype was significantly lower than that in luminal A and luminal B subtypes. By contrast, human epidermal growth factor receptor-2 (HER-2)–overexpressing subtype had higher ¹⁸F-FDG uptake than the other 3 subtypes. There were 8 breast cancer lesions with higher ¹⁸F-alfatide II uptake than ¹⁸F-FDG uptake, which all had a common characteristic that HER-2 expression was negative and estrogen receptor expression was strongly positive. Conclusion: ¹⁸F-alfatide II is suitable for clinical use in breast cancer patients. ¹⁸F-alfatide II is of good performance, but not superior to ¹⁸F-FDG in identifying breast cancer. ¹⁸F-alfatide II may have superiority to ¹⁸F-FDG in detecting breast cancer with strongly positive estrogen receptor expression and negative HER-2 expression.

Angiogenesis exerts a prominent role in promoting tumor growth, progression, and metastasis. Integrin α_vβ₃ is highly expressed on activated endothelial cells of tumor neovasculature and thus is key to tumor angiogenesis (1–3). Arginine-glycine-aspartate (RGD) peptides have a high binding affinity with integrin α_vβ₃. As a result, a variety of RGD-based molecular probes have been developed to visualize integrin α_vβ₃ expression (4–6). Because of the superiority of PET molecular imaging technique, RGD tracers labeled with positron-emitting radionuclides such as ¹⁸F, ⁶⁴Cu, ⁶⁸Ga, and ⁸⁹Zr have attracted much attention (7–10). As a PET tracer based on dimeric RGD peptide, ¹⁸F-alfatide II has been recently proven to possess excellent clinical translational potential in several studies (11–16). Consequently, it warrants further promotion in clinical applications.

Breast cancer is a heterogeneous disease, and angiogenesis is one of the important characteristics (17). Great strides have been made in breast cancer treatment, such as endocrine therapy, targeted human epidermal growth factor receptor-2 (HER-2) therapy, and antiangiogenic therapy. Bevacizumab, as an antiangiogenic drug, was approved by the U.S. Food and Drug Administration to treat patients with advanced breast cancer in 2008. Three years later, however, this approval was withdrawn due to a lack of evidence in improving overall survival. The unsuccessful predicament of bevacizumab could be attributed to the absence of prescreening to select for patients with specific angiogenic biomarkers (18). Accordingly, imaging angiogenesis is crucial for breast cancer patients before antiangiogenic therapy. Moreover, the present molecular classification of breast cancer is based on the status of estrogen receptor (ER), progesterone receptor (PR), and HER-2, which are all expressed on the membrane of breast cancer cells themselves. The biomarkers located in breast cancer stroma are not included in the present classification. With continuous development and improvement of the molecular classification system, some stromal biomarkers such as integrin α_vβ₃ might be incorporated into new classifications in the future. Consequently, the molecular imaging visualization of integrin α_vβ₃ expression should be valuable in exploring the molecular classification based on breast cancer stroma.

There have been a few studies using RGD-based PET probes such as ¹⁸F-galacto-RGD, ¹⁸F-AH111585, ¹⁸F-FPPRGD2, and ⁶⁸Ga-PRGD2 in the clinic for breast cancer patients (19–24). Generally, these PET tracers have shown satisfactory performance in terms of safety, feasibility, and usefulness. However, the numbers of breast cancer cases included in these studies are often very small. Some diagnostic parameters such as sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) are unknown for RGD-based PET tracers due to the absence of benign breast lesions in these studies. Moreover, ductal carcinoma in situ (DCIS) cases were not included in these investigations. For this reason, the information concerning RGD uptake in early breast cancer patients remains unclear. Furthermore, there have been inadequate studies regarding RGD uptake by different subtypes based on the present molecular classification of breast cancer.

In this work, ¹⁸F-alfatide II was investigated in clinically suspected breast cancer patients for the first time, and it was meanwhile compared with ¹⁸F-FDG. We obtained diagnostic parameters of ¹⁸F-alfatide II facilitating the differentiation between breast cancer and benign breast lesions and further assessed its diagnostic performance in comparison with ¹⁸F-FDG. We also investigated ¹⁸F-alfatide II uptake in different molecular subtypes of breast cancer and evaluated the differences as compared with ¹⁸F-FDG.

MATERIALS AND METHODS

Patients

This study was approved by the ethics committee of Jinling Hospital (approval no. 2015NZYW-007) and registered at ClinicalTrials.gov (NCT02582801). Inclusion criteria consisted of clinically suspected primary breast cancer according to conventional imaging (e.g., mammography, ultrasound, MRI), no prior treatment for breast lesions, 18 y < age < 70 y. Exclusion criteria consisted of pregnancy, lactation period, and some accompanied serious diseases (e.g., impaired liver or renal function, active tuberculosis, other malignant tumor). Forty-four female patients (age range, 28–66 y; mean age ± SD, 50.73 ± 8.01 y) were included in this study and each patient signed a written informed consent form. This was a preliminary clinical study about the diagnosis of breast cancer, and a study about treatment response monitoring is ongoing in a larger group of patients.

PET/CT Acquisition and Image Analysis

¹⁸F-alfatide II was prepared according to the previously reported method (25). The injected activity was 306 ± 80 MBq (range, 155–503 MBq). No specific patient preparation such as fasting was requested for ¹⁸F-alfatide II PET/CT scanning, which was performed at 60 min after the injection using a Biography 16 PET/CT scanner (Siemens Healthcare). CT acquisition was initially performed with 120 kV, 140 mA, and a slice thickness of 5 mm. Immediately after CT acquisition, PET emission scanning was performed with an acquisition time of 3 min for each bed. PET data were corrected for attenuation using the coregistered CT data and PET images and reconstructed using an iterative algorithm. PET, CT, and fused images were displayed on a Siemens/Syngo user interface.

¹⁸F-FDG PET/CT was performed within 1–7 d before or after ¹⁸F-alfatide II PET/CT. All patients fasted for at least 6 h before receiving an intravenous injection of ¹⁸F-FDG (∼3.7 MBq/kg of body weight). Blood glucose was measured before injection to ensure that the level was below 140 mg/dL. The same procedure was used for ¹⁸F-FDG PET/CT data acquisition as was used for ¹⁸F-alfatide II PET/CT using the same scanner. The uptake time between ¹⁸F-FDG injection and PET/CT acquisition was also 60 min.

The images from 2 PET/CT scans were visually interpreted by a consensus of 2 experienced nuclear medicine physicians, who were masked to the histologic diagnosis and other imaging results. The SUV_max and the SUV_mean were semiquantitatively measured by drawing regions of interest over the breast lesions.

Statistical Analysis

The data were presented as the mean ± SD. The difference between 2 groups was analyzed by Student t test, and differences among 3 or more groups were determined by ANOVA. Receiver-operating-characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance. The area under the curve (AUC) and the cutoff value were further determined at the point with the highest Youden index. The sensitivity, specificity, PPV, and NPV were calculated to compare the differences in diagnostic accuracy. A Pearson correlation coefficient test was performed to determine the correlation between ¹⁸F-alfatide II and ¹⁸F-FDG groups. Statistical analysis was performed using SPSS software (version 17.0; SPSS, Inc.), and a P value of less than 0.05 was considered statistically significant.

RESULTS

Performance of ¹⁸F-Alfatide II in Diagnosis of Breast Cancer

All patients received biopsy or surgery after undergoing 2 PET/CT scans. Fifty-three breast lesions were confirmed by histopathology in 44 patients. Among them, 42 lesions were malignant and the others were benign. The malignant lesions included DCIS (4 cases) and invasive carcinoma (38 cases). The latter was divided into luminal A subtype (6 cases), luminal B subtype (15 cases), HER-2–overexpressing subtype (5 cases), and triple-negative subtype (12 cases). In luminal B subtype, 6 cases were HER-2–positive and 9 cases were HER-2 –negative. Luminal B subtype with positive HER-2 expression is completely different from HER-2–overexpressing subtype (non-luminal subtype). The former is defined as ER- and (or) PR-positive, HER-2–positive, and Ki-67 at any level. The latter is defined as both ER- and PR-negative and HER-2–positive. The benign lesions were composed of breast fibroadenoma (4 cases), breast adenosis (4 cases), breast intraduct papilloma (2 cases), and mastitis (1 case).

Most breast cancer lesions had high uptake (Fig. 1A). Several benign breast lesions also showed increased uptake (Fig. 1B). The SUV_max and SUV_mean of breast lesions are shown in Table 1. In general, breast cancer lesions had higher ¹⁸F-alfatide II uptake than benign breast lesions (P < 0.05). Also, ¹⁸F-FDG accumulation in breast cancer lesions was more intense than that in benign breast lesions (P < 0.05). However, the difference of ¹⁸F-alfatide II SUV between breast cancer and benign lesions was less significant as compared with that of ¹⁸F-FDG SUV (Figs. 1C and 1D).

FIGURE 1.

(A) A 50-y-old patient with HER-2–overexpressing breast cancer (blue arrow), and axillary lymph node metastasis (red arrow) showing high ¹⁸F-alfatide II and ¹⁸F-FDG uptake. (B) A 47-y-old patient with 2 lesions of breast intraduct papilloma (blue arrows) showing high ¹⁸F-alfatide II and ¹⁸F-FDG uptake. Difference of SUVs between breast cancer and benign breast lesion for ¹⁸F-alfatide II (C) and ¹⁸F-FDG (D). *P < 0.05.

TABLE 1.

Comparisons of ¹⁸F-Alfatide II and ¹⁸F-FDG Uptake Between Breast Cancer and Benign Breast Lesion

		18F-alfatide II		18F-FDG
Disease		SUVmax	SUVmean	SUVmax	SUVmean
BC		3.77 ± 1.78	2.25 ± 0.98	7.37 ± 4.48	4.54 ± 2.82
	P value	P < 0.05	P < 0.05	P < 0.05	P < 0.05
BBL		2.37 ± 1.62	1.50 ± 0.92	2.88 ± 2.77	1.75 ± 1.50

As shown in Figure 2, ROC curves of ¹⁸F-alfatide II are located at the upper left of the chance line, just as those of ¹⁸F-FDG, indicating their strong potential for identifying breast cancer. However, the AUC of ¹⁸F-alfatide II is slightly less than that of ¹⁸F-FDG. The corresponding statistics of the ROC curves are shown in Table 2. The maximum Youden index of ¹⁸F-alfatide II SUV_max (52.1%) is lower than that of ¹⁸F-FDG SUV_max (60.4%), but the gap is small. Moreover, the indices are very close between ¹⁸F-alfatide II SUV_mean (56.5%) and ¹⁸F-FDG SUV_mean (55.6%). The cutoff value with a maximum Youden index is 1.6 for ¹⁸F-alfatide II SUV_max, 1.28 for ¹⁸F-alfatide II SUV_mean, 3.68 for ¹⁸F-FDG SUV_max, and 2.26 for ¹⁸F-FDG SUV_mean. When the SUV_max or SUV_mean of the breast lesion is higher than its cutoff value, breast cancer may be considered. However, ¹⁸F-FDG SUV_mean has the largest AUC (0.84) and ¹⁸F-alfatide II SUV_max has the smallest AUC (0.738).

FIGURE 2.

ROC curves of SUVs of ¹⁸F-alfatide II and ¹⁸F-FDG in differentiating breast cancer from benign lesion.

TABLE 2.

ROC Quantitative Analysis of ¹⁸F-Alfatide II and ¹⁸F-FDG in Differentiating Breast Cancer from Benign Breast Lesion

Tracer	Cutoff value	AUC	Sensitivity	Specificity	Youden index
18F-alfatide II	SUVmax 1.6	0.738	97.6%	54.5%	52.1%
	SUVmean 1.28	0.752	92.9%	63.6%	56.5%
18F-FDG	SUVmax 3.68	0.838	78.6%	81.8%	60.4%
	SUVmean 2.26	0.840	73.8%	81.8%	55.6%

The visual analysis showed that both ¹⁸F-alfatide II and ¹⁸F-FDG for differentiating breast cancer from benign breast lesions had high sensitivity (88.1% vs. 90.5%) and PPV (88.1% vs. 88.4%). However, they had the same specificity (54.5%) and similar NPV (54.5% vs. 60%), which were moderate. The false-positive lesions were the same for ¹⁸F-alfatide II and ¹⁸F-FDG, including 2 cases of breast fibroadenoma (Fig. 3), 1 case of breast adenosis, 1 case of breast intraduct papilloma, and 1 case of mastitis. One case of DCIS was false-negative for both ¹⁸F-alfatide II and ¹⁸F-FDG. Other false-negative lesions of ¹⁸F-alfatide II included 4 cases of breast cancer with triple-negative subtype (Fig. 4). Two cases of breast cancer with luminal B subtype (HER-2–negative) and 1 case of invasive lobular carcinoma was also false-negative for ¹⁸F-FDG. As a result, the sensitivity and NPV significantly increased to 97.6% and 85.7%, respectively, by combining ¹⁸F-alfatide II and ¹⁸F-FDG. However, there was no change on the specificity and only a slight increase on PPV (Table 3).

FIGURE 3.

A 50-y-old patient with breast fibroadenoma (blue arrows) showing false-positive uptake of both ¹⁸F-alfatide II (SUV_max: 5.65) and ¹⁸F-FDG (SUV_max: 3.57).

FIGURE 4.

A 46-y-old patient with triple-negative breast cancer (blue arrows) and axillary lymph node metastasis (red arrows) showing no increased ¹⁸F-alfatide II uptake but intense ¹⁸F-FDG uptake.

TABLE 3.

Visual Analysis of ¹⁸F-Alfatide II and ¹⁸F-FDG in Differentiating Breast Cancer from Benign Breast Lesion

Tracer	Sensitivity	Specificity	PPV	NPV
18F-alfatide II	88.1%	54.5%	88.1%	54.5%
18F-FDG	90.5%	54.5%	88.4%	60.0%
18F-alfatide II + 18F-FDG	97.6%	54.5%	89.1%	85.7%

Comparisons Between ¹⁸F-Alfatide II and ¹⁸F-FDG Uptake in Breast Cancer and Benign Breast Lesions

The ¹⁸F-alfatide II and ¹⁸F-FDG uptake values in breast cancer and benign breast lesions are listed in Table 4. The uptake of ¹⁸F-alfatide II was significantly lower than that of ¹⁸F-FDG in breast cancer lesions (P < 0.05), for both SUV_max and SUV_mean. ¹⁸F-alfatide II uptake in benign breast lesions was also lower than ¹⁸F-FDG, but the difference was not statistically significant (P > 0.05). The correlation between ¹⁸F-alfatide II and ¹⁸F-FDG uptake in breast cancer lesions was not significant (r = 0.03, P > 0.05 for SUV_max and r = 0.003, P > 0.05 for SUV_mean). There was a positive correlation between ¹⁸F-alfatide II and ¹⁸F-FDG uptake in benign breast lesions, which was present for SUV_max (r = 0.69, P < 0.05) but absent for SUV_mean (r = 0.59, P > 0.05) (Fig. 5).

TABLE 4.

Comparisons of ¹⁸F-Alfatide II and ¹⁸F-FDG Uptake in Breast Cancer and Benign Breast Lesion

		BC		BBL
Tracer		SUVmax	SUVmean	SUVmax	SUVmean
18F-alfatide II		3.77 ± 1.78	2.25 ± 0.98	2.37 ± 1.62	1.50 ± 0.92
	P value	P < 0.05	P < 0.05	P > 0.05	P > 0.05
18F-FDG		7.37 ± 4.48	4.54 ± 2.82	2.88 ± 2.77	1.75 ± 1.50

FIGURE 5.

Correlation between ¹⁸F-alfatide II uptake and ¹⁸F-FDG uptake, respectively, based on SUV_max and SUV_mean in breast cancer (A and B) and benign breast lesions (C and D).

¹⁸F-Alfatide II Uptake in Different Molecular Subtypes of Breast Cancer

The ¹⁸F-alfatide II and ¹⁸F-FDG uptake values in different molecular subtypes of breast cancer are shown in Table 5. Luminal A, luminal B, and HER-2–overexpressing subtypes have similar ¹⁸F-alfatide II uptake values, with all being higher than those of triple-negative subtype. According to statistical analysis, there was significant difference in ¹⁸F-alfatide II uptake between triple-negative subtype and luminal A subtype, as well as between triple-negative subtype and luminal B subtype (P < 0.05). No other 2 subtypes of breast cancer had significant difference in ¹⁸F-alfatide II uptake (P > 0.05) (Table 6). In comparison, ¹⁸F-FDG uptake of HER-2–overexpressing subtypes was significantly higher than that of the other 3 subtypes (P < 0.05). Moreover, ¹⁸F-FDG uptake in triple-negative subtype was significantly higher than that in luminal B subtype (P < 0.05). There was no significant difference in ¹⁸F-FDG uptake between luminal A subtype and luminal B subtype, or between luminal A subtype and triple-negative subtype (P > 0.05) (Table 7).

TABLE 5.

¹⁸F-Alfatide II and ¹⁸F-FDG Uptake in Different Molecular Subtypes of Breast Cancer

	18F-alfatide II		18F-FDG
Molecular subtype	SUVmax	SUVmean	SUVmax	SUVmean
Luminal A	4.72 ± 2.56	2.87 ± 1.42	6.08 ± 3.75	3.77 ± 2.30
Luminal B	4.35 ± 1.72	2.62 ± 1.10	5.36 ± 2.64	3.22 ± 1.63
HER-2 overexpressing	4.41 ± 1.92	2.61 ± 1.03	13.88 ± 3.89	8.61 ± 2.23
Triple negative	2.83 ± 0.82	1.77 ± 0.53	9.30 ± 4.11	5.80 ± 2.69

TABLE 6.

P Values of Multiple Comparisons Using ¹⁸F-Alfatide II SUV_max Among Different Molecular Subtypes of Breast Cancer

Molecular subtype	Luminal A	Luminal B	HER-2 positive	Triple negative
Luminal A	—	0.658	0.763	0.032
Luminal B	0.658	—	0.951	0.026
HER-2 overexpressing	0.763	0.951	—	0.088
Triple negative	0.032	0.026	0.088	—

TABLE 7.

P Values of Multiple Comparisons Using ¹⁸F-FDG SUV_max Among Different Molecular Subtypes of Breast Cancer

Molecular subtype	Luminal A	Luminal B	HER-2 positive	Triple negative
Luminal A	—	0.67	0.001	0.074
Luminal B	0.670	—	0.000	0.006
HER-2 overexpressing	0.001	0.000	—	0.019
Triple negative	0.074	0.006	0.019	—

Eight breast cancer lesions had higher ¹⁸F-alfatide II uptake than ¹⁸F-FDG. Among them, 6 lesions were luminal B subtype (HER-2–negative), which accounted for 66.7% (6/9) in all cases of this subtype (Fig. 6A). The remaining 2 lesions were luminal A subtype, with a proportion of 33.3% (2/6) of this subtype (Fig. 6B). Moreover, negative HER-2 expression and strongly positive (3+) ER expression were together shown in the 8 breast cancer lesions. The PR expression was diverse in these lesions.

FIGURE 6.

(A) A 60-y-old patient with luminal B (HER-2–negative) breast cancer (blue arrows) and axillary lymph node metastasis (red arrows) showing intense ¹⁸F-alfatide II uptake but no ¹⁸F-FDG uptake. (B) A 54-y-old patient with luminal A (HER-2–negative) breast cancer (blue arrows) showing increased ¹⁸F-alfatide II uptake, but no ¹⁸F-FDG uptake, and distant metastases with larger numbers and more intense uptake (green arrows) by ¹⁸F-alfatide II than by ¹⁸F-FDG.

DISCUSSION

RGD-based PET tracers have undergone rapid development in recent years. ¹⁸F-galacto-RGD was the first noninvasive probe to target integrin α_vβ₃ for PET angiogenesis imaging. Like ¹⁸F-Galacto-RGD, ¹⁸F-AH111585 is another monomeric RGD PET tracer. There have been several preclinical and clinical studies suggesting the usefulness of these 2 PET tracers (20,22,26–30). However, they have relatively low tumor uptake, because of limited RGD binding of monomeric peptides and rapid clearance of the peptide tracers. Accordingly, radionuclide-labeled RGD dimers, including ¹⁸F-FPPRGD2 and ⁶⁸Ga-PRGD2, have been further developed to achieve better performance (31–33). The use of a fluoride-aluminum complex, ¹⁸F-Al-NOTA-PRGD2 (denoted as ¹⁸F-alfatide), significantly simplifies the labeling procedure (34–37).

For the first time, ¹⁸F-alfatide II was clinically used in differentiating breast cancer from benign breast lesions in our study. As compared with previous studies using other RGD-based PET tracers in breast cancer patients (19,20,22–24), our study included more diverse cases, such as breast fibroadenoma, breast adenosis, DCIS, invasive carcinoma, and lobular carcinoma. Even though there were 5 cases of early cancer in our study, breast cancer overall still showed relatively high ¹⁸F-alfatide II uptake with a SUV_max of 3.77 ± 1.78 and SUV_mean of 2.25 ± 0.98, higher than that in benign breast lesions, which demonstrated that ¹⁸F-alfatide II is suitable for identification of breast cancer in clinical practice.

In comparison with ¹⁸F-FDG, ¹⁸F-alfatide II had less difference in uptake between breast cancer and benign lesions. AUC of ¹⁸F-alfatide II was also lower than that of ¹⁸F-FDG in diagnosis of breast cancer. Some diagnostic parameters such as Youden index (SUV_max), sensitivity, PPV, and NPV for ¹⁸F-alfatide II were slightly lower than those for ¹⁸F-FDG, whereas the specificity was same. These results indicated that ¹⁸F-alfatide II has a diagnostic value comparable to that of ¹⁸F-FDG but is not superior in identification of breast cancer.

In our study, 5 cases of benign breast lesions displayed false-positive uptake for both ¹⁸F-alfatide II and ¹⁸F-FDG, suggesting that ¹⁸F-alfatide II does not help improve the detection specificity. As for lobular carcinoma, it has been known that the false-negative rate of ¹⁸F-FDG is as high as 65.2% (38). One case of lobular carcinoma included in our investigation also showed no ¹⁸F-FDG uptake but had intense ¹⁸F-alfatide II uptake (Fig. 4B). This phenomenon was consistent with that reported in previous studies using ¹⁸F-FPPRGD2 (19), indicating that RGD-based PET tracers such as ¹⁸F-alfatide II can be complementary to ¹⁸F-FDG in diagnosis of lobular carcinoma of the breast. Other false-negative cases of ¹⁸F-FDG included 1 case of DCIS and 2 cases of breast cancer with luminal B subtype (negative HER-2 expression). The former case was also deficient in ¹⁸F-alfatide II uptake because of the small lesion size. The latter 2 cases, however, showed increased ¹⁸F-alfatide II uptake, suggesting the additive role of ¹⁸F-alfatide II to ¹⁸F-FDG in detecting this subtype of breast cancer. On the other hand, 4 cases of triple-negative subtype without ¹⁸F-alfatide II foci showed significantly increased ¹⁸F-FDG uptake, demonstrating that ¹⁸F-FDG can compensate for the deficiency of ¹⁸F-alfatide II in this subtype. Accordingly, combining ¹⁸F-alfatide II and ¹⁸F-FDG together can significantly improve the sensitivity and NPV but may not add much in terms of the specificity and PPV.

Comparisons between ¹⁸F-alfatide II and ¹⁸F-FDG uptake were also made in our study. The result showed that tumor uptake of ¹⁸F-alfatide II was significantly lower than that of ¹⁸F-FDG in breast cancer. This finding was similar to those presented in other RGD tracer studies (19,21). It has been reported in several studies that there was no significant correlation between RGD-based PET tracers and ¹⁸F-FDG in cancer lesion uptake (21,24). Likewise, our result revealed a lack of significant correlation between ¹⁸F-alfatide II uptake and ¹⁸F-FDG uptake in breast cancer.

¹⁸F-alfatide II uptake in different molecular subtypes of breast cancer was also assessed in our study. The triple-negative subtype showed no or low ¹⁸F-alfatide II uptake, which was significantly lower than that of luminal A and luminal B subtypes. By contrast, the triple-negative subtype showed high ¹⁸F-FDG uptake, whereas luminal A and luminal B subtypes had relatively low ¹⁸F-FDG uptake. This result was similar to that from the previous study using ⁶⁸Ga-PRGD2 in breast cancer patients (23).

Furthermore, our study evaluated the breast cancer lesions with higher uptake of RGD peptide than that of ¹⁸F-FDG for the first time. A total of 8 such lesions were found in our study. Interestingly, they shared a common feature of being HER-2–negative and strongly ER-positive (3+). This finding may help guide the therapeutic direction of breast cancer patients with multiple metastases, which are not classified by conventional molecular subtypes. If these metastases have higher uptake of ¹⁸F-alfatide II than that of ¹⁸F-FDG, they are very likely to be strongly ER-positive and HER-2–negative. Accordingly, they could benefit from endocrine therapy such as tamoxifen, but not from anti–HER-2 therapy such as trastuzumab.

There exist some limitations in our study. First, the number of participants is not large enough. Second, lymph node and other distant metastases were not evaluated due to the limited number of metastatic lesions in these patients. Third, immunohistochemistry tests were not performed to assess the correlation between integrin α_vβ₃ expression and alfatide II uptake, which has been demonstrated in several animal and clinical studies (12,16,39). Further investigations are still required in the future to elucidate the role of ¹⁸F-alfatide II in breast cancer management.

CONCLUSION

¹⁸F-alfatide II is clinically amenable for the identification of breast cancer without special patient preparation. ¹⁸F-alfatide II has good diagnostic value in distinguishing between breast cancer and benign breast lesions, but is not superior to ¹⁸F-FDG. ¹⁸F-alfatide II can exert a complementary role to ¹⁸F-FDG in breast lobular carcinoma and other breast cancers with strongly positive ER expression and negative HER-2 expression.

DISCLOSURE

This work was supported by the National Key Basic Research Program of China (973 program, 2014CB744504), the National Natural Science Foundation of China (81501537), the Natural Science Foundation of Jiangsu Province (BK20160610), Jiangsu Province Social Development Program (BE2017772), Jiangsu Planned Projects for Postdoctoral Research Funds (1601090C), and the Intramural Research Program, National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health. Jihong Tian, Chuanjin Sun, and Xingang Wang also contributed to this work. No other potential conflict of interest relevant to this article was reported.