Comparing [18F]FDG-positron emission tomography and breast magnetic resonance imaging to predict pathological complete response and 3-year invasive disease-free survival in HER2-positive early breast cancer patients: an unplanned exploratory analysis of the PHERGain trial.

BACKGROUND

The PHERGain trial demonstrated that an [18F]2-fluoro-2-deoxy-d-glucose ([18F]FDG)-positron emission tomography (PET)-based, pathological complete response (pCR)-adapted strategy could be safely utilized to avoid chemotherapy (CT) in patients with human epidermal growth factor receptor 2 (HER2)-positive early breast cancer (EBC) receiving neoadjuvant dual anti-HER2 blockade [trastuzumab and pertuzumab (HP)]. Due to the limited availability of [18F]FDG-PET, this study evaluated breast magnetic resonance imaging (MRI) as an alternative for early treatment response assessment.

PATIENTS AND

METHODS

Group B patients (n = 285) initially received two cycles of HP, with subsequent CT introduction if [18F]FDG-PET showed no response. [18F]FDG-PET and MRI were conducted before randomization and after two cycles (early). An additional MRI was carried out before surgery (late). Concordance between [18F]FDG-PET, MRI reduction, and MRI response by RECIST v.1.1 was evaluated, along with accuracy to predict pCR and 3-year invasive disease-free survival (iDFS) rates.

RESULTS

Early imaging assessment showed good accuracy (78.2%) between [18F]FDG-PET and breast MRI tumor reduction (any shrinkage), but not when applying RECIST v.1.1 response criteria (≥30% decrease in the sum of diameters of target lesions). There were higher pCR rates in [18F]FDG-PET responders with early MRI reduction (39.0% versus 29.6% if no reduction) or MRI response (44.0% versus 30.4% if no response). [18F]FDG-PET non-responders without MRI reduction had the lowest pCR (21.7%) and 3-year iDFS (75.3%) rates despite receiving CT. Among [18F]FDG-PET responders, early MRI complete responses (CRs) were uncommon, but extending CT-free treatment increased early MRI CR (9.3%-31.7%) and objective response rates (55.1%-70.0%). Late MRI CR predicted pCR better in hormone receptor (HR)-negative than in HR-positive tumors (positive predictive value: 85.5% versus 61.5%).

CONCLUSIONS

Although [18F]FDG-PET is the recommended imaging technique for guiding treatment in HER2-positive EBC patients following the PHERGain strategy, this unplanned analysis suggests that tumor shrinkage assessed by breast MRI could be a viable alternative for adaptive strategies in settings where [18F]FDG-PET is not available.