Inclusion criteria: Patients whose histopathology results were breast carcinoma. They were diagnosed in HOH from June 2019 to June 2021. Their stage was evaluated by standard modalities: breast MRI; breast, neck and abdominal ultrasound, mammography, and bone scan. All of these patients then underwent ¹⁸F-FDG PET/CT and were re-evaluated post-scan. Patients officially agreed to participate in the study.

Exclusion criteria: Pregnant women; patients with severe comorbidities (heart failure, kidney failure); Patients whose medical records were incomplete; patients already underwent chemotherapy and/or radiotherapy; Patients who refused to participate the study.

Radiopharmaceutical: ¹⁸F-FDG (¹⁸F-fluoro-2-deoxyglucose).

Equipment: Whole-body PET/CT imaging was performed using PET/CT scanner system (GE Discovery Light Speed).

Study design: descriptive study using retrospective and prospective data.

Sampling: convenient sampling—all patients who satisfied inclusion criteria and did not have any exclusion criterion were selected.

Procedure:

- All diagnostic data were collected, including breast MRI; breast, neck and abdominal ultrasound, mammography, bone scan.

- ¹⁸F-FDG PET/CT imaging performance: Patients were fasting for at least 6 hours before being injected with ¹⁸F-FDG, 0.15 mCi/kg. All patients took contrast agents orally, and no IV contrast agent was used. PET/CT scan was performed 60 minutes after ¹⁸F-FDG injection, from the vertex of the skull to the mid femoral region, with extra scans on focused areas when necessary. PET/CT images were evaluated by at least 2 nuclear medicine specialists.

Image processing and result evaluation: the quantity, location, size of the primary tumor, metastatic regional lymph node(s), distant metastatic lesion(s), images of ¹⁸F-FDG uptake and distribution, and SUV were evaluated. SUV of detected lesions were compared to the radioactivity of the mediastinal blood pool (SUV mean value = 2.5). If the lesion had SUV from 2 to 2.5, it was suspected between benign and malignant. If its SUV was above 2.5, its malignancy was confirmed.

Data analysis: Data were processed by SPSS 20.0.

Among 51 patients qualified for the study, there were 46 agreed to participate in. All were females, with the average age of 56.6 years old (range 33 - 85 years old). Results showed that the majority of patients were from 40 to 60 years old, accounted for 58.7%, whereas the young age group (less than 40 years old) had the lowest proportion.

The most common location of the primary tumor was the upper outer quadrant of the breast (61%). The tumors in the upper inner quadrant accounted for 28%, and those in the lower outer accounted for 7%. The least common location of the primary tumors was the lower inner quadrant, appearing in only 4% of patients.

According to Table 1 and Figure 2, SUVmax mean value of primary tumors was 6.39, with standard deviation (SD) = 4.02. There was a positive correlation between tumor size and SUVmax with r = 0.759. That means, the larger the tumor size was, the higher the SUV was, with p value = 0.0001.

According to Table 2, there were 18 patients underwent surgery. Those patients had taken pre-operative PET/CT and histopathology after PET/CT. Eight of them had malignant lymph node(s) on their histopathology results, while the rest (10) had benign result. Conventionally, if the SUVmax is above 2.5, the lymph node is considered malignant. Among 8 patients in the malignant group confirmed by histopathology, 6 cases had SUVmax above 2.5, thus, the sensitivity of ¹⁸F-FDG-PET/CT in lymph node evaluation was 75% (6/8). Among 10 patients having the benign result confirmed by histopathology, all of them had SUVmax below 2.5, so the specificity ¹⁸F-FDG-PET/CT in lymph node evaluation was 100% (10/10) in our study.

¹⁸F-FDG PET/CT detected distant metastases on 7/46 breast carcinoma patients (15.2%), which consisted of 2 bone metastasis cases, 1 lung metastasis case, 1 liver and bone metastasis case, 1 bone and distant lymph node metastasis case, and 1 distant lymph node metastasis case.

As presented in Table 3, four patients had bone metastasis, in which there was a total of 11 metastatic lesions, found in spinal bone, pelvis bone, scapula, and femur. Totally, there were 26 distant metastatic lesions detected on ¹⁸F-FDG PET/CT. There was a statistically significant difference among ¹⁸F-FDG SUVmax mean values of metastatic sites, with the highest mean value of SUVmax found in bone metastasis.

According to Table 4, there were 21/46 patients (45.6%) had their TNM stages changed. Especially, distant metastases were detected in 7 patients, whereas the TNM stage of 1 patient was changed from end stage to operable stage. Consequently, the initial treatment strategies of these 8 patients (17.4%) have been altered.

* p < 0.001; ** There were 2 patients having multiple-site metastases.

^aChanged from lower stage to higher stage; ^bChanged from higher stage to lower stage; ^cDistant metastasis detected by ¹⁸F-FDG PET/CT; ^dChanged from end stage to earlier, operable stage.

¹⁸FDG uptake level and tumor metabolism varied among patients. In our study, the mean ¹⁸FDG SUVmax was 6.39 ± 4.02, with the highest value recorded being 17.8. Comparing SUVmax among tumors, there was a statistically significant difference in tumor size corresponding to tumor stage (<2 cm, 2 - 5 cm, and >5 cm). There was a positive correlation between tumor size and SUVmax. This is a relatively strong correlation, with r = 0.759. That means, the larger the tumor size was, the higher the SUV value was, with p = 0.0001.

In our study, primary tumors were found in the left breast more frequently than in the right breast (54.3% vs. 45.7%). This finding was similar to that of other national studies. In Vu Kien’s study on 116 breast cancer patients, the rate of left breast cancer was higher than that of the right one (54% vs. 46%) [5]. Our study also revealed that the most common primary tumor location was the upper outer quadrant of the breast (61%), which is also similar to the result of other studies. According to Le Hong Quang (2012), the rate of tumors detected in the upper outer quadrant was the highest, accounted for 51.2% cases [6]. Nguyen Thi Sang

et al. (2016) also found that the upper outer quadrant of the breast is the most common location of primary tumors (39.9%), followed by the lower outer quadrant (26.5%) [7].

Internationally, McMasters et al. studied 2148 patients and detected 51.0% of primary tumors on the upper outer quadrant [8]. This finding could be explained by the anatomic structure of the breast. Mammary glands tend to extend towards the axillary fossa, therefore the quantity of glands and the volume of breast tissues in the upper outer quadrant is the largest among all locations. Hence, the rate of breast carcinoma occurring in this location is highest.

The size of the primary tumor and that of axillary lymph node(s) are the most important prognostic factors, which significantly impact the five-year survival rate of breast cancer patients. In our study, the average tumor size was 2.8 cm, ranging from 1.1 cm to 7.8 cm. Most of our research subjects were at early stages and under pre-operative evaluation, therefore the majority have tumor size at the T2 stage.

According to the study of Sang Kyu Yang et al., the ability to evaluate breast tumors on PET/CT depends on the tumor size and its histopathological characteristics. Tumor sensitivity was recorded at 68% for tumors under 2 cm, and at 92% for tumors from 2 - 5 cm. In case of tumor in situ (Tis) with low sensitivity (2% - 25%), the ability of PET/CT to detect breast tumors is low if tumors are small or if they are non-invasive breast cancer [9] .

¹⁸FDG uptake level and tumor metabolism varied among patients. In our study, the mean ¹⁸FDG SUVmax was 6.39 ± 4.02, with the highest value recorded being 17.8. Comparing SUVmax among tumors, there was a statistically significant difference in tumor size corresponding to tumor stage (<2 cm, 2 - 5 cm, and >5 cm). There was a positive correlation between tumor size and SUVmax. This is a relatively strong correlation, with r = 0.759. That means, the larger the tumor size was, the higher the SUV value was, with p = 0.0001.

This finding is similar to that of Song’s study on 55 patients, with the mean SUVmax being 6.3 ± 4.8. SUVmax of primary tumors varied among stages, and the difference was statistically significant, which means that SUVmax increased along with tumor size. Patients with high SUVmax had their disease advanced faster, compared to those with lower SUVmax [10]. Hence, assessing ¹⁸F-FDG uptake level on PET/CT could help prognose for patients.

In breast cancer, determining axillary lymph node metastasis is crucial for treatment planning and prognosis. Axillary lymph node dissection for histopathology is the most accurate method to detect lymph node metastasis. However, its complications such as lymphedema, swollen axillary skin, or reduced movement of the shoulder would eventually lead to a decrease in patient’s quality of life. Thus, accurately staging lymph nodes by non-invasive methods is necessary to help patients avoid unnecessary axillary node dissection.

In our study, the sensitivity of ¹⁸F-FDG-PET/CT in lymph node evaluation was 75% (6/8), and its specificity was 100% (10/10). Wahl et al. studied 360 breast cancer patients who underwent ¹⁸F-FDG PET/CT and concluded that the sensitivity of ¹⁸F-FDG-PET/CT in axillary lymph node evaluation was 61% and the specificity was 81% [11]. Furthermore, ¹⁸F-FDG PET/CT was proven to be more efficient in evaluating supraclavicular, infraclavicular, and mediastinal lymph nodes. According to Schirrmeister et al., the sensitivity and specificity of ¹⁸F-FDG-PET/CT in axillary lymph node detection were 79% and 92% respectively [12]. PET/CT could accurately detect the anatomic location and distinguish metastatic and non-metastatic lymph nodes when visible on CT [9]. ¹⁸F-FDG PET/CT shows superiority in lymph node evaluation over conventional diagnostic imaging modalities.

In our study, distant metastasis was found in 7/46 patients (15.2%), with a total of 26 distant metastasis lesions detected on ¹⁸F-FDG-PET/CT including 2 bone metastasis cases, 1 lung metastasis case, 1 liver and bone metastasis case, 1 bone and distant lymph node metastasis case, and 1 distant lymph node metastasis case. All of these patients were diagnosed with stage II-III breast cancer before PET/CT.

Groheux et al. took PET/CT on 117 breast cancer patients, and compared PET/CT staging with staging by other, conventional modalities including bone scan, abdominal ultrasound, chest X-ray, and chest CT. On PET/CT, authors detected distant metastases in 43 patients, whereas other diagnostic imaging modalities detected metastases in only 28 cases [13].

In comparison, PET/CT is a better modality for detecting metastases in bone, liver, and lymph nodes than other conventional diagnostic imaging modalities. PET could effectively detect mediastinal lymph nodes, even those under 1 cm. When both PET and CT of ¹⁸F-FDG PET/CT have detected bone metastases, the bone scan is no longer necessary. An advantage of PET/CT compared to other diagnostic imaging methods such as chest X-ray, bone scan, or abdominal ultrasound, is the ability to identify distant metastatic lesions in various organs in only one examination [9].

Among 26 metastatic lesions (4 lung lesions, 11 bone lesions, 4 liver lesions, and 7 distant lymph node lesions) in our study, the mean value of ¹⁸F-FDG SUVmax of metastatic bone was the highest (9.26 ± 3.49), followed by that of distant metastasis lymph node and liver, which were 7.54 ± 2.87 and 7.37 ± 0.93 respectively, whereas metastatic lesions in lung has the lowest mean value of ¹⁸F-FDG SUVmax. There was a statistically significant difference among ¹⁸F-FDG SUVmax of different metastatic lesion groups by site (p < 0.001). Cokmert et al. performed pre-treatment ¹⁸F-FDG PET/CT on 176 breast cancer patients with distant metastasis. Regression analysis results showed that SUVmax below 7.55 could be an independent indicator having prognostic value for survival rate in all stage IV cancer patients [14]. Hence, our study and previous studies have shown the superior ability of ¹⁸F-FDG PET/CT in detecting distant metastases, compared to other conventional diagnostic imaging modalities.

Staging plays the key role in planning treatment strategies for most cancer patients. Based on domestic and international studies, Ministry of Health of Vietnam has established “Guidelines for breast cancer diagnosis and treatment”. Specifically, for stage I and II breast cancer, the initial treatment method is surgery, followed by chemotherapy, radiotherapy, or hormone therapy. Stage III breast cancer is divided into two groups: instantly operable and uninstantly operable diseases. In case of instantly operable diseases, the recommended approach is surgery first, followed by chemotherapy, WBRT (Whole Breast Radiotherapy), RNI (Regional Nodal Irradiation), and hormone therapy when HR (Hormone Receptor) is positive. In case of tumors adhesive to the chess wall, and/or adhesive and immobile axillary lymph nodes, the recommended indications would be chemotherapy first with 6 - 8 circles (depends on the regimen), then surgery. If the disease responds to chemotherapy and downstages to be operable, surgery would be performed, followed by radiotherapy, and hormone therapy if HR is positive. If the disease does not respond to chemotherapy, a different regimen of radiotherapy could be considered to reduce the tumor’s size and invasive level. In case of overexposed HER-2/neu receptor disease, trastuzumab—a monoclonal antibody—could be used. For stage IV patients, the treatment plan would include chemotherapy, hormone therapy, and molecular biology therapy [15].

Comparing stages before and after ¹⁸F-FDG PET/CT, we found that ¹⁸F-FDG PET/CT could detect meaningful lesions, which alter completely the disease stage, such as metastatic lesions in supraclavicular lymph nodes, internal mammary lymph nodes, bone, lung, liver, and distant lymph nodes. Among 46 patients, there were 7 patients had their stages changed to stage IV because distant metastases were detected on their PET/CT scan, whereas in another patient, the stage was down from IV to II (operable disease). Consequently, these 8 patients had their treatment strategy changed after PET/CT, accounted for 17.5% cases.

This finding is consistent with that of the study of Koolen et al. (2012). In that study, they took ¹⁸F-FDG PET/CT on 154 stage IIB - III breast cancer patients. Distant metastatic lesions were found in 13% of patients, which changed their stages to stage IV [16]. In another study conducted by Groheux et al., during a period of 71 months, they performed ¹⁸F-FDG PET/CT scan for 254 stage II - III breast cancer patients (their definitive diagnosis determined by clinical examination, mammography, breast MRI and ultrasound). Results showed that there were 77 patients had stage changed (30.3%, p = 0.05). N3 nodes were detected in 40 patients, and distant metastases were found in 53 cases. ¹⁸F-FDG PET/CT detected distant metastases in 1/44 stage IIa patients (2.3%), 6/56 stage IIb patients (10.7%), 11/63 IIIa patients (17.5%), and 27/74 stage IIIb patients (36.5%) [13].

In another study, Garami et al. performed ¹⁸F-FDG PET/CT scan on 115 breast cancer patients, in whom no regional lymph node, distant metastasis, or primary tumor size below 4 cm were detected by other diagnostic imaging modalities. The sensitivity in detecting primary tumors of ¹⁸F-FDG PET/CT was 93%. The sensitivity and specificity of lymph node ultrasound were 30% and 95% respectively, while those of ¹⁸F-FDG PET/CT were 72% and 96%. ¹⁸F-FDG PET/CT also detected distant metastases in 8 patients, changed the stage of 54 patients (47%), and altered the treatment plan of 18 patients (15.6%). Thus, ¹⁸F-FDG PET/CT could detect distant metastases in 7% - 8% of patients whose metastatic lesions were previously unable to be detected by any other diagnostic imaging modalities [17].

Overall, our findings are consistent with and confirm the findings of other international studies on the advantages of ¹⁸F-FDG PET/CT over other, conventional diagnostic imaging modalities in evaluating accurately the breast carcinoma stage before treatment, which help determine most suitable treatment strategies for patients.

However, our study still has limitations. First, our small sample size and convenient sampling could lead to not very high representative value of the data. It is because in a developing country as Vietnam, the accessibility to PET/CT is still limited due to unavailability of equipment and its high expense that makes it unaffordable for many patients. Second, we couldn’t conduct further analysis because patients came from different provinces where the database was separated, which made collecting data and monitoring patients more difficult. To overcome these limitations, it is recommended to conduct further studies with larger sample size and more comprehensive methodology and analysis. Additionally, an upgrade in the database is necessary, and a radiopharmaceutical manufacturing system needs to be developed to lower the cost of PET/CT to make it more accessible with majority of cancer patients.