Abstract Background: In HER2+ breast cancer (BC), residual disease (RD) after neoadjuvant HER2 inhibition is associated with increased risk of recurrence. However, the genetic and non-genetic resistance mechanisms underlying RD remain incompletely characterized. Here we used single nuclei (sn) RNA and DNA sequencing to profile genetic and transcriptional changes in growing, residual, and recurrent tumors from an inducible mouse tumor model of HER2+ breast cancer. Methods: We used mammary tumors/tumor bed (n=2-3/time point) from female MMTV-rtTA; TetO-ERBB2 mice at 9 weeks DoxOn (HER2on), 7 days DoxOff (HER2off) and 12-21 weeks Doxoff (recurrent). Samples were snap frozen and single nuclei isolation was performed. snRNAseq was performed using 10 × 5’ v2 kits for library preparation and sequencing. snDNAseq was performed using a custom amplicon panel targeting the mouse genome and human ERBB2 locus via the MissionBio Tapestri platform. Results: For snRNAseq analysis, tumor cell clusters were identified by Epcam and ERBB2 gene expression. Tumor cells were further subclustered into 18 unique clusters that separated by ERBB2 expression. Seven subclusters of HER2on tumor cells were characterized by low ERRB2 expression: 1.38 vs 2.94 (median normalized log counts) in HER2on-ERBB2high cells. Top 5 differentially expressed genes (DEG) per condition (log2 fold change >1.5, adjusted p value < 0.05) were: HER2on: Erbb2, Csn3, Lcn2, Csn2, Wfdc18; HER2on-ERBB2high: Csn3, Igfbp5, Csn2, Lcn2, Trf; HER2on-ERBB2low: Lars2, Gm47283, Cmss1, Gm42418, Neat1; HER2off: Lars2, Hexb, Cmss1, Gphn, Gm42418; Recurrent: Capza2, Vim, Lgals1, Crabp1, S100a6. HER2on-ERBB2low cells were more closely related in transcriptional space to HER2off residual tumor cells than HER2on-ERBB2high tumor cells. Trajectory analysis suggested that HER2off RD evolved from HER2on-ERRB2low cells and formed the basis of recurrent disease. To determine clonal evolution over time, we used common top 10 DEGs between HER2on vs HER2off vs HER2low snRNAseq data and intersected that set with significantly enriched single nucleotide variants (SNVs) from our snDNAseq data. We identified 16 clones containing at least 1 SNV at >1% frequency across all samples. We observed clonal persistence, relative enrichment and extinguishment between HER2on, HER2off and recurrent samples but without clonal dominance. Conclusion: Time-series analysis of genetic and transcriptional evolution in a mouse model of HER2+ BC RD suggests that a tumor cell subpopulation characterized by an ERBB2low-associated transcriptional program drives RD after HER2 inhibition. Citation Format: Sheheryar K. Kabraji, Weihua Wang, Yali Zhang, Johann Bergholz, Tao Jiang, Vittal Kurisetty, Maia Homsi, Jianmin Wang, Jean Zhao. Residual disease after HER2 inhibition is driven by a primary tumor subpopulation expressing an ERBB2-low associated transcriptional program [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6939.