Correlation of Maximum Standardized Uptake Values in 18F-Fluorodeoxyglucose Positron Emission Tomography-computed Tomography Scan with Immunohistochemistry and Other Prognostic Factors in Breast Cancer.

OBJECTIVES: The objective of this study was to correlate maximum standardized uptake value (SUVmax) with different immunohistochemical subtypes of breast cancer and other prognostic factors in breast cancer. SUBJECTS AND METHODS: This was a retrospective study including 219 consecutive patients undergoing whole-body fluorodeoxyglucose positron emission tomography/computed tomography scan for the staging of breast cancer. Out of 219 patients, two were male and 217 were female; age ranged from 26 to 85 years with mean age of 54 years. On histopathological examination (HPE), 197 patients were of invasive ductal carcinoma type and two of lobular type. Histopathological grades, immunohistochemistry (IHC) types, and ki-67 values were compared with SUVmax values.
RESULTS: The mean SUVmax of the population was 11.39 (±6.05). The mean SUVmax in different HPE grades was Grade 1 = 6.81 ± 5.6, Grade 2 = 11.4 ± 6.12, and Grade 3 = 13.14 ± 5. The mean SUVmax values in different IHC types were Luminal A = 7.75 ± 4.2, Luminal B = 10.01 ± 5.3, triple negative = 15.26 ± 5.6, and HER2 enriched = 11.27 ± 5.2. The mean SUVmax in high ki-67 patients was 11.97 ± 5.85 compared with 7.25 ± 3.43 patients with low ki-67. Univariate analysis showed significant difference in SUVmax in patients with different grades (P = 0.013), hormone receptor positivity (P ≤ 0.001), ki-67 (P < 0.001), and axillary lymph node positivity (P ≤ 0.001). In multivariate regression analysis, there was significantly higher SUVmax value in triple-negative patients after correcting for tumor size, ki-67 value, axillary lymph node status, and grade of tumor.
CONCLUSION: High SUVmax values were noted in high-grade, high ki-67, triple-negative, and axillary lymph node positive tumors.

Introduction

Breast cancer is the most common cancer and second leading cause of cancer-related deaths in women.[1] Breast cancer constitutes 30% of new cancer cases diagnosed in women. Majority of breast cancer cases present in women above the age of 40 years and only 7% of breast cancers develop in <40 years.[2] Accurate staging and prognostication is very essential for proper management. Breast cancer is a very heterogeneous tumor with prognosis depending on various histopathological and immunohistochemical factors. Major prognostic factors considered in breast cancer include size of tumor, focality, lymph node spread, distant metastases, and various histopathological and molecular features such as histopathological type, grade, hormone receptor status, HER2neu status, ki-67 value, P-53 tumor suppressor gene expression, and cerb2 proto-oncogene expression. St. Galen International consensus report in 2011, classified cancer into four types based on immunohistochemistry (IHC): Luminal A, Luminal B, triple negative, HER2neu positive.[3] Triple-negative breast cancer, constituting around 20% of patients, have aggressive histology and poor prognosis compared to other types.[45]

Whole-body fluorodeoxyglucose positron emission tomography-computed tomography (FDG PET/CT) scan is a very useful imaging modality in staging of breast cancers.[67] FDG uptake in tumor cells is based on Warburg effect.[8] Higher FDG uptake is noted in tumors with aggressive biology and tends to have poorer prognosis.[6] The purpose of this study is to correlate FDG uptake in primary breast cancer with various prognostic factors.

Subjects and Methods

This was a retrospective study of 219 patients who underwent whole-body FDG PET/CT scan at our institute for the staging of breast cancer between January 2016 and November 2017. All the patients were histopathologically proven. Age of the patients ranged from 26 to 85 years with mean age of 54.07 ± 11.83 years. Patients have been variously categorized according to age, tumor size, histopathological type, grade, IHC, and ki-67 status. Based on age, patients are divided into two groups: <40 and ≥40 years. Based on tumor size, patients are divided into two groups: <2 cm and ≥2 cm. Totally 213 patients underwent IHC. Totally 197 patients underwent ki-67 status of tumor. Estrogen and progesterone receptors status was considered as positive with 10% as cutoff. Patients were divided into four IHC types: Luminal A, Luminal B, triple negative, and HER2neu positive. ki-67 <14% was considered as low and ≥14% as high.

Whole-body FDG PET/CT scan images were acquired from the vertex of the skull to mid-thigh on GE discovery STE scanner with 16 slices CT. Before injecting 5.18 MBq/kg of 18F-FDG, the patient fasted for minimum of 6 h and blood sugar levels were <150 mg/dl. Patients were instructed to avoid muscular activity. Three dimensional PET acquisition was done with 3 min/bed. CT transmissions maps were used for attenuation correction. PET images were reconstructed using optimum subset expectation maximization algorithm. Images were displayed and interpreted in ADW 4/4.5 workstations. The region of interest was drawn around the lesions manually. FDG PET/CT scans were interpreted by two experienced nuclear medicine physicians. Maximum standardized uptake value (SUVmax) value of primary breast mass was measured.

Mean SUVmax values (±standard deviation) were calculated in different subgroups as described. Univariate regression analysis was performed using SPSS Statistics for Windows, Version 21.0. IBM Corp; Armonk, NY to find out any significant difference in SUVmax in the subgroups. The subgroups having significant difference were included in the multivariate regression analysis. P < 0.05 was considered statistically significant.

Results

The age of the patients ranged from 26 to 85 years with mean age of 54.07 ± 11.83. SUVmax ranged from 1.4 to 37.7 with mean SUVmax of 11.39 ± 6.05. The mean SUVmax values for various subgroups have been depicted in Table 1.

Table 1

Depicting mean maximum standardized uptake value values in various subgroups along with number of patients in various groups along with univariate and multivariate analysis results

Sample number and SUVmax values in various sub groups			P
Parameter	n	SUV (mean±SD)	Univariate analysis	Multivariate analysis
Age			0.13
<40	32	12.87±6.26
≥40	187	11.1±0.13
Grade			0.013
1	10	6.81±5.67
2	188	11.4±6.1
3	21	13.14±5.2
Histology
Invasive ductal type	197	11.387±6.13
Invasive lobular type	2	5.25±0.21
Ductal with lobular differentiation	1	9.3
Others	19	11.85±6.0
Hormone receptor status			0.000
Positive	105	13.37±5.7
Negative	114	9.5±5.8
HER2neu status			0.648
Positive	149	11.49±6.5
Negative	114	11.08±5.05
Ki67 status			0.000	0.011
High	167	11.97±5.8
Low	30	7.25±3.4
Axillary lymph nodes			0.000	0.002
Positive	165	12.5±5.9
Negative	54	7.8±5.3
Distant metastases			0.086
Present	91	12.2±5.7
Absent	128	10.76±6.2
Tumor size (cm)			0.000	0.002
<2	44	7.95±4.5
≥2	175	12.21±6.16
IHC types			0.000
Luminal A	23	7.75±4.30	0.001
Luminal B	90	10.01±5.34	0.003
Triple negative	55	15.26±5.69	0.000	0.000
HER2 Enriched	45	11.27±5.20	0.880

SUVmax: Maximum standardized uptake value, IHC: Immunohistochemistry, SD: Standard deviation

A univariate analysis was performed in these same set of subgroups [Table 1]. A statistically significant difference was noted in the following subgroups: grade of tumor, different IHC types, hormone receptor status, ki-67 status, size of tumor, and axillary node status. Triple-negative patients have significantly higher SUVmax value compared to patients with other IHC subtypes. On multivariate regression analysis [Table 1], there was significantly higher SUVmax value in triple-negative patients after adjusting for tumor size, histopathology grade, high ki-67, and axillary lymph node status (P < 0.0001).

Discussion

Breast cancer is a very heterogeneous type of tumor in terms of pathology, tumor biology, and clinical response to therapy. Preoperative prognostication is very important as tumors with poor prognosis can be down staged with neoadjuvant chemotherapy, and breast conservative surgery may be planned. There are several well-established prognostic factors in breast cancer which include tumor grade, pathological type, IHC type based on receptor status, HER2neu status and ki-67 levels, axillary lymph node status, and distant metastases.

FDG PET/CT scan is useful in staging[791011] early response assessment,[12] restaging, and prognostication of breast cancer.[1314] FDG uptake in tumor can be expressed by simple quantitative parameter like SUVmax. In current study, metabolism of breast cancer was correlated with tumor size, histopathology grade, high ki-67, triple-negative IHC, and axillary lymph node metastases. These results are similar to the previously published studies.[151617181920] Since most of the tumors were of ductal type (n = 197) in our study, a correlation between SUVmax and different histological types was not performed.

There is statistically significant difference in SUVmax values in patients with different IHC types with Luminal A having least value followed by Luminal B, HER2 enriched, and highest in triple-negative patients [Figure 1]. Among Luminal B patients, those with HER2 positivity have higher mean SUVmax compared to those with HER2 negativity (11.1 vs. 9.6), although sample number in each group is not enough to have sufficient statistical power. Has Şimşek et al.[21] in their retrospective study that included 436 patients showed that there was significant difference in SUVmax value between HER2-positive and-negative patients. In the present study, there was no significant difference in SUVmax value between HER2-positive and -negative patients. The apparent difference in study by Has Şimşek et al. may be related to other confounding prognostic factors. Mean SUVmax values of different IHC types in the present study were comparable to the values in study by Has Şimşek et al.

Figure 1

Positron emission tomography-computed tomography scan images with maximum standardized uptake value values in different immunohistochemistry types

Basu et al.[22] in their study that included 62 patients (18 triple negative and 44 nontriple negative) found that there was significantly higher SUVmax value in triple-negative tumors compared to nontriple negative tumors. In the present study, also triple-negative tumors have significantly higher tumor metabolism compared to nontriple negative tumors. Axillary lymph node metastases are considered as the most important prognostic factor in breast cancer. In our present study, patients with positive axillary lymph nodes showed higher SUVmax value compared to patients with negative axillary lymph nodes.

Although some studies show a correlation between SUVmax value and distant metastases,[7] the present study does not reveal any statistically significant difference. Since this is a retrospective study, it may be subjected to selection bias.

Conclusion

SUVmax value in FDG PET/CT scan is independently associated with large tumor size, triple-negative status, high ki-67, and axillary lymph node positivity.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.