Among the top causes of female mortality, breast cancer remains a major global health concern (1, 2). According to projections by the American Cancer Society, approximately 313,510 individuals in the United States are expected to be diagnosed with breast cancer in 2024, while an estimated 42,780 are anticipated to succumb to the disease (3). In the 10 years following initial treatment, at least 35% of breast cancer patients experience local or distant recurrence (4). Breast cancer frequently recurs locally and metastasizes distantly, leading to poor prognosis and high mortality. Therefore, it is essential for patients to get a timely diagnosis of local recurrences and distant metastases.

Breast cancer recurrence is generally evaluated with a physical exam, as well as a mammogram, computed tomography (CT), magnetic resonance imaging (MRI), and bone scintigram. Despite detecting masses and lymph node abnormalities, physical examinations lack sensitivity to detect small metastases and early recurrences, complicating accurate staging (5). Both CT and MRI are susceptible to false positives due to postoperative fibrosis and scarring, and their resolution may be insufficient for small lesions (6). Although bone scans are highly sensitive for detecting bone metastases but they lack specificity, which can result in misdiagnoses due to non-cancerous conditions (7). These methods frequently struggle to distinguish between postoperative changes and true pathology, especially in early or small recurrences.

To address these limitations, recent advancements in multimodal and molecular imaging have introduced new approaches for detecting breast cancer recurrence and metastasis. [¹⁸F] Fluorodeoxyglucose positron emission tomography/CT ([¹⁸F]FDG PET/CT) and [¹⁸F]FDG PET/MRI are rapidly advancing molecular imaging technologies widely used for detecting and localizing breast cancer recurrence. [¹⁸F]FDG is a radiotracer that determines cellular metabolism by exploiting in vivo metabolic properties, which enables it to detect tumors and abnormal tissues (8). [¹⁸F]FDG PET/CT combines positron emission tomography and computed tomography, offering precise metabolic data and high-resolution anatomical images, crucial for identifying and analyzing recurrent lesions (9). [¹⁸F]FDG PET/MRI integrates positron emission tomography with magnetic resonance imaging, providing both metabolic and functional information, potentially advantageous for studying the biological traits of recurrent lesions (10). In addition, PET/MR excels in soft tissue contrast, making it more sensitive to vascular and soft tissue diseases (11). However, the relative diagnostic performance of these technologies remain debated.

In this study, articles with [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI results have been evaluated and compared to detect recurrent breast cancer. A direct comparison in this context has not been thoroughly explored, making this study an important contribution to the understanding of imaging techniques in breast cancer recurrence evaluation.

This meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses for Diagnostic Test Accuracy (PRISMA-DTA) guidelines for diagnostic test accuracy (12). PROSPERO network (ID CRD42024556069) prospectively registered the protocol.

Current guidelines from prominent oncology organizations, including the American Society of Clinical Oncology, the National Comprehensive Cancer Network, and the European Society for Medical Oncology, advise that follow-up care for breast cancer patients who are asymptomatic should primarily consist of regular physical examinations and annual mammographic screenings. The use of additional laboratory or imaging tests on a routine basis is not recommended within these guidelines (36–38). However, for breast cancer patients with a suspected recurrence, there is currently no standardized follow-up protocol in place, making additional radiologic imaging crucial, particularly when recurrence is suggested by elevated tumor markers or concerning symptoms.

A prospective study evaluated [¹⁸F]FDG PET/CT in 100 patients with suspected breast cancer recurrence compared to thoracoabdominal contrast-enhanced CTs (ceCT) and bone scans (39). The findings revealed that 22% of patients were diagnosed with distant recurrence, 19% had local recurrence only, and 59% exhibited no evidence of recurrence. The diagnostic accuracy for identifying distant recurrence was assessed using the area under the ROC curve, with [¹⁸F]FDG PET/CT achieving an AUC of 0.99, compared to 0.84 for ceCT and 0.86 for the combination of ceCT and bone scans (39). Another study reported that [¹⁸F]FDG PET/CT exhibits a high positive predictive value of 0.97 and an accuracy ranging from 0.83 to 0.86 in detecting suspected recurrent condition (24, 40). Additionally, other clinical guidelines suggest that [¹⁸F]FDG PET/CT can be particularly useful in identifying the site of recurrence when conventional imaging methods yield inconclusive results (41). In contrast, another meta-analysis has demonstrated that [¹⁸F]FDG PET/MRI offers robust diagnostic accuracy, with a sensitivity of 0.94 and specificity of 0.90 for nodal staging, as well as a sensitivity of 0.98 and specificity of 0.96 for distant staging (42). Preliminary data suggest that PET/MRI could be a viable alternative to PET/CT for current clinical applications (43). In conclusion, both [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI perform well in detecting breast cancer recurrence. However, a systematic comparison is still lacking, leaving the question of which modality is superior unresolved. This meta-analysis seeks to fill a gap in the literature by comparing the diagnostic sensitivity and specificity of [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI in breast cancer recurrence.

The meta-analytic outcomes of this systematic review demonstrate that there is no statistically significant difference in overall sensitivity between [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI for the detection of breast cancer recurrence at both the lesion level (p = 0.71) and the patient level (p = 0.07). Similarly, the overall specificity of the two imaging modalities does not differ significantly at the lesion level (p = 0.44) or the patient level (p = 0.06). These findings indicate that both imaging techniques provide comparable diagnostic performance in clinical settings. These findings collectively indicate that both imaging techniques demonstrate equivalent diagnostic accuracy in the detection of recurrent breast cancer within clinical practice.

The diagnostic equivalency observed can be attributed to the common use of [¹⁸F]FDG, a glucose analog that preferentially accumulates in metabolically active tumor cells (44). While PET/MRI provides enhanced soft tissue contrast, especially in breast and pelvic areas, the diagnostic efficacy is predominantly determined by the metabolic imaging aspect (45, 46). Consequently, the PET-based assessment of glucose metabolism is likely the primary determinant of sensitivity and specificity in both imaging modalities.

In addition, at the lesion-based analysis, a study assessing the performance of these imaging modalities revealed that [¹⁸F]FDG PET/MRI correctly identified a greater proportion of lesions, achieving a sensitivity of 98.5% compared to 94.8% for [¹⁸F]FDG PET/CT (32). This finding suggests that PET/MRI may be more effective in detecting smaller or less conspicuous lesions that could be overlooked by PET/CT (32). Furthermore, patient-based analyses indicate that [¹⁸F]FDG PET/MRI offers advantages over PET/CT. A meta-analysis comparing these modalities found that [¹⁸F]FDG PET/MRI exhibited higher sensitivity for detecting distant metastases in breast cancer patients, with a sensitivity of 1.00 compared to 0.96 for PET/CT (47).

To date, no comprehensive meta-analysis or extensive comparative study has specifically investigated the influence of recurrence site on the relative diagnostic efficacy of [¹⁸F]FDG PET/CT versus PET/MRI. This omission constitutes a significant gap in the existing literature, as elucidating lesion-specific diagnostic advantages could facilitate the development of more personalized and anatomically targeted imaging strategies for patients with suspected recurrence. In response to this gap, the current study is pioneering in its examination of the impact of recurrence site on imaging modality performance through stratified subgroup analyses, providing novel insights into how anatomical context may affect diagnostic sensitivity and specificity. In our meta-analysis, both analytical methodologies produced consistent results, demonstrating no statistically significant difference in sensitivity or specificity between [¹⁸F]FDG PET/CT and PET/MRI at either the lesion level or the patient level. This internal consistency enhances confidence in the robustness of our findings and indicates that the diagnostic equivalence between these modalities is maintained irrespective of the level of analytical granularity. However, future research should clearly differentiate between these analytical levels and, where feasible, employ hierarchical modeling approaches that account for the clustering of lesions within patients.

Furthermore, this meta-analysis offers a thorough assessment of the diagnostic effectiveness of [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI for detecting breast cancer recurrence, filling critical gaps from earlier research. Xiao et al. (48) conducted a systematic review and meta-analysis, demonstrating the high sensitivity (0.90) and specificity (0.81) of [¹⁸F]FDG PET/CT in detecting breast cancer recurrence. While their findings highlighted the diagnostic value of [¹⁸F]FDG PET/CT, they did not compare it with PET/MRI, leaving unresolved the question of which modality performs better in this context. Our study builds upon their work by providing the first head-to-head comparison between PET/CT and PET/MRI, offering a clearer, evidence-based understanding of the diagnostic capabilities of both modalities in breast cancer recurrence.

In contrast, Dan et al. (49) focused on the diagnostic performance of PET/MRI in breast cancer, including initial diagnosis, lymph node involvement, and bone metastasis. While their study demonstrated superior sensitivity (0.95) and specificity (0.94) of PET/MRI for breast lesions and metastases, a direct comparison between PET/MRI and PET/CT was not conducted, leaving an important gap in understanding their relative performance.

In comparing the diagnostic efficacy of [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI in breast cancer recurrence, both modalities demonstrated similar sensitivity and specificity, indicating comparable performance in detecting recurrent disease. Nonetheless, each option has its unique benefits and drawbacks. Several studies have demonstrated that PET/CT is a cost-effective and widely available option, including one emphasizing its utility for monitoring metastatic breast cancer treatment responses (50). PET/CT also offers quicker scan times, reducing patient discomfort and rendering it a viable option in numerous clinical practice. A PET/MRI scan, on the other hand, enhances the contrast of soft tissue while reducing radiation exposure, which is especially important for younger or more vulnerable patients (51). In light of these factors, while PET/CT appears to be more economical and accessible, PET/MRI’s enhanced diagnostic performance and safety profile in specific cases cannot be ignored. The complementary strengths of these modalities suggest that combining them into a hybrid diagnostic approach may improve overall diagnostic accuracy. Patients’ clinical circumstances and diagnostic needs should ultimately determine whether PET/CT or PET/MRI is appropriate.

Interpreting the results of this meta-analysis requires consideration of certain limitations. First, the heterogeneity among the included studies may have influenced the pooled sensitivities and specificities of [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI. In order to identify potential sources of heterogeneity and evaluate the robustness of our findings, we conducted a leave-one-out sensitivity analysis. The results demonstrated stability across all iterations, with no individual study exerting a disproportionate influence on the aggregated sensitivity or specificity estimates. This consistency indicates that our meta-analytic conclusions are robust and not significantly impacted by any outlier study. Moreover, the imbalance in studies between [¹⁸F]FDG PET/CT (16 studies) and [¹⁸F]FDG PET/MRI (4 studies) may introduce bias, with PET/CT’s larger sample size potentially affecting statistical power and confidence intervals for PET/MRI. We conducted subgroup analyses by anatomical regions and sample sizes to address this, but the disparity still limits our analysis’s robustness. Future research should aim for more balanced and larger sample sizes for both imaging methods to ensure more reliable results. Another key limitation is the reliance on retrospective studies (9 out of 16), which are susceptible to selection and verification biases, potentially impacting our findings. Although prospective cohort studies would be preferable, the unpredictable nature of breast cancer recurrence makes them impractical. Thus, we used retrospective data despite its limitations. Recognizing these biases is crucial, and future research should prioritize prospective, multi-center cohort studies to minimize biases and improve result generalizability. The absence of standardized imaging protocols across studies is a limitation, as variations in MRI sequences and PET methods may lead to inconsistent diagnostic performance. A universally accepted protocol for imaging breast cancer recurrence is lacking. Developing such standardized protocols should be a research priority to enhance result consistency and enable reliable comparisons between imaging modalities. Lastly, while we assessed publication bias using a funnel plot and Egger’s test, it’s crucial to recognize that studies favoring PET/MRI may dominate the literature, potentially inflating its perceived diagnostic efficacy. Research on breast cancer recurrence often emphasizes imaging techniques like PET/MRI, leading to a higher likelihood of publishing positive results. This bias can misrepresent PET/MRI’s true effectiveness, as studies with negative or inconclusive outcomes may be underreported. Future research should aim to include both positive and negative findings to provide a more accurate evaluation of PET/MRI’s clinical value in detecting breast cancer recurrence.

To overcome the discussed limitations, we suggest larger, balanced studies, standardized imaging protocols, detailed site comparisons, and reducing publication bias. These measures are crucial for enhancing the diagnostic accuracy of [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI in detecting breast cancer recurrence, thereby improving oncology clinical decisions.

The findings of our meta-analysis suggest that [¹⁸F]FDG PET/CT and [¹⁸F]FDG PET/MRI exhibit comparable sensitivity and specificity in diagnosing breast cancer recurrence at both the lesion and patient levels. In a clinical setting, both of these imaging techniques have their respective strengths and limitations, and physicians should take these into account when making the most suitable choice for patients. Considering that the number of articles on [¹⁸F]FDG PET/MRI in diagnosing recurrence of breast cancer is fewer compared to [¹⁸F]FDG PET/CT, more head-to-head researches are needed to expand this field to obtain more robust and comprehensive results.