See the associated article on page 1212.
Metastatic breast cancer from an estrogen receptor (ER)–positive primary tumor is rarely cured, but patients often live for many years with their disease (1). A wide range of therapy regimens are available, including endocrine therapy, cytotoxic chemotherapy, and molecularly targeted agents. Without established guidelines, clinicians and patients are looking for biomarkers to direct sequencing or combine these therapies. Metastatic disease may have vastly different characteristics compared with a treated primary tumor, but contemporaneous biopsies may yield inadequate tissue (2) and may not represent the patient’s full tumor burden.
In this issue of The Journal of Nuclear Medicine, Nienhuis et al. (3) demonstrate the potential contribution of molecular imaging to assessment of metastatic breast cancer, as they document 18F-fluoroestradiol (18F-FES) SUVmax (SUV of the hottest pixel) for 1,617 lesions in 91 patients. Nienhuis et al. interpret their results (shown graphically in Fig. 1 (3)) as indicating that 36% of patients have site-to-site heterogeneity of disease, with both 18F-FES–positive and 18F-FES–negative lesions. With the application of agglomerative hierarchical cluster analysis to imaging-detected disease characteristics (e.g., number of 18F-FES–positive lesions, percentage of 18F-FES–positive lesions, average 18F-FES SUVmax, number of bone lesions, number of lung lesions; Supplemental Fig. 2 of Nienhuis et al. (3)), the 91 patients are partitioned into 3 groups primarily based on tumor 18F-FES avidity, number of tumors, and tumor location. These results and small differences by lesion type in average geometric mean 18F-FES uptake led the authors to conclude that “18F-FES uptake is heterogeneous between tumor lesions … and is influenced by anatomic site.” The article provides valuable data on an important topic, but further consideration is required to determine the role of 18F-FES PET/CT imaging in metastatic breast cancer.
The first potential role of 18F-FES PET/CT is to address the clinical dilemma of treatment selection and sequencing for metastatic breast cancer. The Nienhuis et al. study suggests several clinical predictions, such as that patients with any 18F-FES–negative lesion are unlikely to respond to endocrine therapy, and that patients with visceral disease are unlikely to respond to endocrine therapy. These straightforward hypotheses were evaluated but not strongly supported in a similar patient population (4). In that study, clustering based on tumor aggressiveness (measured by 18F-FDG uptake) and average 18F-FES uptake was robust to internal cross-validation and identified 3 groups with median progression-free survival ranging from 3.3 to 26.1 mo. The clustering described by Nienhuis et al. (3) could have more clinical impact if additional clinical features were considered, such as prior exposure to different therapy types and time between primary and metastatic diagnosis. In general, biomarkers should be assessed in the context of standard prognostic variables (5). The authors also propose background (normal tissue) correction for normalization of tumor 18F-FES uptake measures, but this would require additional reader effort and add another source of measurement error. Studies EAI142 (NCT02398773) and IMPACT-MBC (NCT01957332) are ongoing to observe relationships between 18F-FES uptake and response to endocrine therapy, but prospective biomarker-driven trials (6) are required to determine the role of 18F-FES PET measures in clinical practice.
The second potential role of 18F-FES PET/CT is to inform development of new therapies that target the ER and to contribute to research into the mechanisms for development of metastatic disease. 18F-FES PET may be used for pharmacodynamic monitoring of ER blockade in both preclinical (7,8) and clinical (9,10) studies. For broader insights into disease development Nienhuis et al. (3) interpret their results as indicating that site-to-site heterogeneity within patients is an important consideration for metastatic breast cancer therapeutic development. Kurland et al. (11) examined similar lesion-level data and concluded (from patterns of 18F-FES uptake quite comparable to those in Fig. 1 of Nienhuis et al. (3)) that site-to-site heterogeneity could be attributed largely to measurement error and that co-occurrence of lesions with extremely high and extremely low uptake was uncommon. This interpretation was supported by subsequent analysis in which progression-free survival was predicted by patient-level averages rather than characteristics defined by site-to-site heterogeneity (4). Differences in ER expression have been documented to occur between primary and metastatic disease (12,13), among different contemporaneous metastatic sites (14), and intratumorally (15). Understanding clonal evolution in response to multiple lines of treatment is clearly of fundamental interest for metastatic breast cancer, but other sources of information and extensive preclinical studies are required to provide context to the findings of clinical 18F-FES PET/CT. Researchers with expertise in molecular imaging and genomic analyses should coordinate their efforts for optimal discovery.
A part of enabling effective cross-disciplinary collaboration in metastatic breast cancer is better nomenclature for “heterogeneity” to distinguish among patterns with very different implications for clinical practice and basic research. When a group of patients treated as homogeneous by clinical guidelines (metastatic breast cancer from an ER-positive primary tumor) has different average response to endocrine therapy based on a different classifier (such as PET/CT imaging), this indicates that breast cancer is a heterogeneous disease. When this disease heterogeneity is referred to as interpatient heterogeneity, it invites parallels to the unrelated phenomenon of intrapatient heterogeneity, either over time or in synchronous disease. 18F-FES PET imaging has great promise for revealing disease heterogeneity in metastatic breast cancer from an ER-positive primary tumor. Second, site-to-site heterogeneity, different measurements for different tumors within the same person, is also detectable by 18F-FES PET, but the existence of lesions with uptake somewhat above and somewhat below a prespecified threshold does not necessarily yield actionable information. Finally, intratumoral heterogeneity, variability of measures within a single tumor, is of great relevance for understanding tumor biology (16), and at some level can be assessed by PET imaging (17,18).
In summary, the Nienhuis et al. study (3) supports prior findings that 18F-FES PET imaging can help in clinically relevant classification of patients with metastatic breast cancer from an ER-positive primary tumor and presents correlative studies in normal tissue to guide further development of 18F-FES uptake measures. The statement that uptake is influenced by the site of metastasis requires further study to evaluate possible clinical impact or biologic insight; the number of evaluable visceral tumors was relatively small, and low sensitivity of CT to bone lesion identification could lead to an artifactual overrepresentation of bone lesions with higher 18F-FES uptake. We look forward to the future development of 18F-FES imaging and the treatment of metastatic breast cancer.
DISCLOSURE
This work was supported by NIH grant U01-CA148131. No other potential conflict of interest relevant to this article was reported.
Footnotes
Published online Jun. 14, 2018.
- © 2018 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication June 8, 2018.
- Accepted for publication June 13, 2018.