Correlation between 18F-FDG uptake on PET/CT and proliferation index in patients with breast cancer

Objectives Positron emission tomography (PET) imaging using the radiotracer ¹⁸F-Fluorodeoxyglucose (FDG) has been proposed as imaging biomarkers of cell proliferation. If FDG PET/CT can be established as such, it will provide a non-invasive, quantitative measurement of tumor proliferation across the entire tumor. Results from validation studies have so far been conflicting with some studies confirming a good correlation between FDG uptake and Ki-67 score in breast cancer and others presenting negative results. We aimed to explore the correlation of FDG uptake with the Ki-67 labeling index in patients with breast cancer.

Methods The MEDLINE, EMBASE, Cochrane Library and China National Knowledge Infrastructure (CNKI) databases were systematically searched for all relevant literature about the relationship of FDG uptake and Ki-67 expression. The publishing language was limited to English and Chinese. Retrieved articles were screened carefully based on the selection criteria. Then the quality of included studies was evaluated according to the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. The correlation coefficient (r) and its 95% confidence interval (CI) of individual studies were meta-analyzed using a random-effects model. The 95% CIs were calculated by applying Fisher’s z transformation and inverse Fisher transformation. Higher values indicated stronger correlation (r<0.21: poor correlation; 0.21&#8804;r<0.41: average correlation; 0.41&#8804;r<0.61: moderate correlation; 0.61&#8804;r<0.81: significant correlation; r蠅0.81: strong correlation). Additionally, Begg’s test was performed to assess the publication bias, and the sources of heterogeneity were explored by sensitivity and subgroup analysis.

Results Twenty-two articles involving 1139 patients were included for this meta-analysis, comprising 11 studies retrospectively designed and the others prospectively designed. For the scanning modality, most of the included studies used the integrated PET/CT while only 4 studies used PET alone. Almost all studies provided SUVmax measurement as the uptake index. With regard to the measurement of Ki-67 expression, the majority of studies measured the highest proliferation rate (Ki-67_max) in the samples, which were usually obtained by surgical resection. The results of QUADAS-2 showed the common weakness was concentrated in the interpretation of index test and reference standard without blindness. In addition, lack of an explicit description of the time interval between the reference standard (PET scan) and index test (Ki-67 immunohistochemistry) also was a problem for most studies. A total of 17 studies did not provide the time interval, and the other 5 studies were assessed as high risk for a long time interval over 4 weeks. The pooled r values for ¹⁸F-FDG/Ki-67 correlation 0.42 (95% CI, 0.37-0.47), which indicated a moderate correlation, but with a significant heterogeneity (I² = 80.9%, P<0.01) (Figure 1). Sensitivity analysis revealed that a single study contributed no significant influence to the overall estimate. The differences between subgroups of study design were not significant (r=0.41 for prospective studies, r=0.43 for retrospective studies, p>0.05). However, r value of PET subgroup was higher than that of PET/CT subgroup (p<0.05) (Table 1). Due to limited information and small sample size, we did not analyze other factors such as tumor type, Ki-67 labeling index, sample collection methods, which might be potential sources of heterogeneity. In addition, there was no significant publication bias for pooled r value.

Conclusions In patients with breast cancer, ¹⁸F-FDG uptake showed a moderate positive correlation with tumor cell proliferation. However, our results need further validation by larger, prospective studies with improved study design.

• Download figure
• Open in new tab
• Download powerpoint