Nadia Arang, Postdoctoral Fellow at UCSF in the laboratory of Nevan Krogan
“Systematic Proteomic Profiling to Reveal Adaptive Remodeling of the PI3K Interaction Network”.

Protein-protein interaction (PPI) networks are highly dynamic and precisely regulated in response to diverse extrinsic and intrinsic cues. In cancer and other complex diseases, these networks can be extensively rewired to drive oncogenesis and modulate therapeutic responses. Despite advances in mapping global interactomes under basal conditions, comparatively little is known about how targeted inhibition of oncogenic drivers, such as phosphoinositide 3-kinase (PI3K), remodels protein interaction networks in cancer cells. PI3K signaling plays a central role in tumor initiation and progression, driving drug development efforts for nearly three decades. Early generation, broad-spectrum inhibitors suppressed all class 1 PI3K catalytic isoforms (p110α, p110β, p110δ, and p110γ), but their lack of selectivity often led to limited therapeutic windows and dose-limiting toxicities3. More selective agents, such as alpelisib, which preferentially target p110α, have demonstrated clinical benefit, leading to FDA approval for PIK3CA-mutant breast cancer. However, because most PI3K inhibitors are orthosteric, binding to the active site of both mutant and wild-type PI3Kα, they disrupt the PI3K signaling systemically, including in metabolic tissues where PI3Kα is essential for insulin-dependent glucose uptake. This causes side effects, such as rash and hyperglycemia, the latter driving compensatory insulin release that can reactivate PI3K signaling in tumor cells. To overcome these limitations, new mutant-selective inhibitors have been designed to preferentially target or degrade the oncogenic p110α H1047R variant—one of the most prevalent activating mutations in breast cancer. While these agents are more specific, responses remain incomplete and often transient. For example, HER2 and HER2- breast cancer tumors have demonstrated differential responses to p110α-targeted agents in the clinic. One proposed mechanism is that in HER2 breast cancer cells, non-selective PI3K inhibitors initially suppress downstream signaling; however, the presence of catalytically inactive p110α enables compensatory receptor tyrosine kinase (RTK)-dependent complex formation and feedback reactivation. In contrast, mutant selective p110α degraders not only abrogate kinase function but also disrupt these complexes, preventing feedback-driven pathway re-engagement7,8. Notably, degradation has been shown to preferentially target p110α associated with p85β regulatory subunits rather than p85α, which may suggest selective recruitment of p85β–p110α complexes to activated RTKs such as HER2 and HER3 and highlights the importance of specific protein assemblies in controlling pathway output and drug sensitivity. Despite these insights, the specific protein complexes and their dynamic remodeling in response to targeted PI3Kα inhibition remain poorly defined. This underscores a critical gap in our understanding of the molecular mechanisms driving PI3K pathway signaling that could benefit from the application of network-based approaches and highlights an urgent need for novel therapeutic strategies for patients with PIK3CA mutations, including breast cancer. Towards this end and extending the existing CCMI focus on systems approaches to study breast cancer, this project will systematically profile how PI3K-targeted inhibitors rewire protein complexes in order to gain mechanistic insights into the adaptive remodeling of signaling networks that underlie therapeutic response. The central premise of our proposal is that HER2 and HER2- PIK3CA-mutant breast cancer cells have distinct protein interaction landscapes that underlie their differential therapeutic responses. Specifically, we hypothesize that in HER2 tumors, stabilized RTK-driven protein complexes maintain persistent signaling and limit sensitivity to PI3K-targeted therapies. In addition, we expect that AP-MS profiling will reveal the reorganization of downstream signaling assemblies, such as mTOR-containing complexes, which may contribute to on-target toxicities, including hyperglycemia.

Kasturi Nayak, Associate Specialist at UCSF in the laboratory of John Gordan
“Functional and phosphoproteomic mapping of PIK3CA mediated resistance to KRASG12C inhibition in colorectal cancer”.

KRAS is one of the most frequently mutated oncogenes in human cancers, including pancreatic, lung, and colorectal cancer (CRC), where it drives tumorigenesis and resistance to therapy. In CRC, KRAS mutations occur in ~35–45% of cases, most commonly at codon 12 (e.g., G12D, G12V, G12C) and are associated with poor prognosis and resistance to anti-EGFR therapies. Although KRASG12C (~3–4% of CRCs) is targetable with covalent inhibitors such as adagrasib and sotorasib, clinical responses in CRC have been modest compared to other cancers, likely due to tissue-specific signaling rewiring and rapid engagement of compensatory pathways. A major downstream effector of KRAS is PIK3CA, which contributes to feedback resistance mechanisms. While PIK3CA mutations act as primary oncogenic drivers in some cancers, they frequently co-occur with KRAS mutations in colorectal cancer (CRC). Targeting one pathway often triggers compensatory activation of the other—PI3K inhibition can upregulate MAPK signaling and vice versa, leading to rapid therapeutic escape and limited monotherapy durability. The essential roles of these pathways in normal tissues further constrain the therapeutic window, underscoring the need for selective, combination-based strategies. Oncogenic PIK3CA can maintain PI3K-AKT signaling independently of RTK input, bypassing KRAS dependency and reducing KRASG12C inhibitor efficacy. Thus, co-targeting could overcome resistance. The emergence of broad-spectrum KRAS inhibitors targeting multiple G12 alleles has expanded therapeutic options, highlighting the need to better understand intrinsic as well as adaptive resistance mechanisms. To map these resistance networks effectively and develop combination therapy strategies for CRC, functional tools that enable high-resolution mapping are essential. In the Gordan Lab, we investigate resistance to KRASG12C inhibition in CRC, aiming to identify lineage-specific vulnerabilities and rational combination strategies. Importantly, we have developed a Cas12a-based, paralog-specific Kinome screening platform in SW837 KRASG12C mutant CRC cells to identify synthetic lethal interactions obscured by gene family redundancy. By integrating this platform with quantitative phosphoproteomics, we can perform high-resolution mapping of oncogenic signaling dependencies in KRAS/PIK3CA co-mutant contexts. The anticipated findings will establish a mechanistically informed framework for precision combination therapies in KRAS-mutant CRC.

Lauren Clubb, Graduate Student at UCSD in the laboratory of J Silvio Gutkind
“Establishing a Multi-Cancer Atlas of Receptor Tyrosine Kinase Expression in Tumor and Tumor Infiltrating Immune Cells.”

Receptor tyrosine kinases are a major target for cancer therapy; however, their varied expression patterns on cancer, immune, and stromal cells complicate their use as therapeutic targets. Receptor tyrosine kinases (RTKs) are a family of transmembrane receptors with intrinsic tyrosine kinase activity that play crucial roles in cell signaling, regulating processes such as cell proliferation, differentiation, metabolism, migration, and survival. Ligand binding induces receptor dimerization and the phosphorylation of multiple key substrates, concomitant with the autophosphorylation of tyrosine residues within the cytoplasmic domains, which serve as docking sites for a variety of signaling proteins, thereby triggering diverse intracellular cascades. In cancer, RTKs are frequently dysregulated through gene amplification, point mutations, and chromosomal rearrangements that generate oncogenic fusion proteins (e.g., EML4-ALK, NTRK fusions). These alterations drive constitutive activation of oncogenic pathways, with well-characterized examples including EGFR and HER2 in lung and breast cancers, MET in lung and gastric cancers, ALK and ROS1 in lung cancer, RET fusions in thyroid and lung cancer, and FGFR alterations in bladder and cholangiocarcinoma. Given their central role in tumorigenesis, RTKs have become key therapeutic targets, with over 100 small-molecule inhibitors and monoclonal antibodies approved by the FDA. Clinically important agents include EGFR inhibitors (erlotinib, osimertinib) and antibodies (cetuximab), HER2-targeting antibodies (trastuzumab, pertuzumab), ALK inhibitors (crizotinib, alectinib, lorlatinib), MET inhibitors (capmatinib, tepotinib), and multi-targeted TKIs (lenvatinib, cabozantinib). Despite these advances, intrinsic, adaptive, and acquired resistance to RTK-targeted therapies remains a major clinical challenge, underscoring the urgent need for next-generation inhibitors and rational combination strategies. A combination of receptor tyrosine kinase inhibitors (RTKi) and immune checkpoint blockade (ICB) therapies is becoming increasingly common; however, the potential off-target effects of RTKi can compromise the efficacy of ICB. Immune checkpoint blockade (ICB) has revolutionized cancer therapy, enabling durable responses in a subset of patients. Monoclonal antibodies targeting co-inhibitory receptors such as PD-1 and CTLA-4 enhance tumor-specific T cell function by alleviating direct immunosuppression from the cancer and tumor-associated cells. However, for most solid tumors, only 20–40% of patients experience lasting benefit, largely due to additional mechanisms of immune evasion. Oncogenic signaling pathways are emerging as key drivers of immune escape and are hence becoming attractive targets for combination therapies. RTKs like AXL, MET, and VEGFR, for example, are expressed on tumor, stromal, and immune cells and may contribute to T cell exclusion, myeloid-derived suppressor cell (MDSC) recruitment, and alternative macrophage polarization. Combination RTK inhibitor (RTKi) and ICB therapy is under active investigation in cancers such as melanoma, endometrial, gastric, colorectal, and renal cancer. While these combinations can enhance anti-tumor immunity and reduce immunosuppression by tumor-associated cells, they also pose risks of toxicity. Similarly, the repertoire of RTKs expressed in immune cells is still poorly understood, hence compromising the ability to harness the full potential of RTKi combinations without compromising the anti-cancer immune response. The complete elucidation of RTK expression across the tumor microenvironment, including in immunosuppressive myeloid cells, Tregs, fibroblasts, and endothelial cells, as well as in dendritic and cytotoxic (CD4 and CD8) T and NK cells, could inform strategies to maximize therapeutic efficacy and minimize off-target effects. The planned studies will build on the collaborative opportunities with leaders at CCMI (Drs. Natalia Jura and Nevan Krogan) as well as with Dr. Tony Hunter, an adjunct professor of Pharmacology at UCSD and the Salk Institute, who initially discovered that kinases are capable of tyrosine phosphorylation. We propose to build an atlas of RTK expression across all major cancer types and within each stromal and immune cell of the tumor microenvironment to expand our understanding of potential immunomodulatory effects of RTK inhibition in the treatment of cancer, and to harness the full potential of novel RTKi/ICB combination therapies, while making our findings available to the entire scientific community through the release of a web app.