Abstract Purpose: To determine whether FDG PET can expand eligibility in biomarker-selected clinical trials by providing a means to quantitate response in patients with non-assessable disease by RECIST. Experimental Design: SUMMIT (NCT01953926) is a multicenter phase II “basket” trial of the Pan-HER kinase inhibitor, neratinib. Patients had advanced ERBB2 (HER2)-mutant solid tumors, ≥1 measurable lesion, preferably defined unidimensionally by RECIST v1.1, or alternatively metabolically by PET Response Criteria (PRC). The primary aim was to determine the proportion of additional breast cancer patients accrued using PRC who would have otherwise been ineligible based on RECIST criteria alone. The secondary aim was to determine the concordance of response versus non-response between RECIST and PRC. Results: Eighty-one patients with HER2-mutant metastatic breast cancer were accrued; 77 were evaluable for response by RECIST and/or PRC. 63 (82%) were RECIST-evaluable and 14 (18%) were accrued using PRC alone. Bone-only disease (n = 11; 79%) was the most common reason for classification as non-measurable by RECIST. Twenty-nine patients were accrued and followed using both criteria, of which 25 (86%; 95% confidence interval, 68%–96%) were concordant for response versus non-response as defined by RECIST and PRC. Conclusions: PRC allowed patients with non-RECIST measurable disease access to therapy and facilitated more rapid accrual of patients to this trial of a rare biomarker. PRC and RECIST both provided methods of response assessment and were generally concordant. Thus, PRC was useful as a supplement to RECIST criteria. This provides a rationale for including FDG PET measurements in future clinical trials involving rare tumors or rare genomically defined subpopulations of more common cancers.