Open Access Anti-cancer effect of afatinib, dual inhibitor of HER2 and EGFR, on novel mutation HER2 E401G in models of patient-derived cancer

Next-generation sequencing-based multi-gene panel testing is increasingly used to guide targeted therapies across tumor types. HER2 alterations are important therapeutic targets, with basket trials (e.g., MyPathway) showing benefit of trastuzumab plus pertuzumab in HER2-amplified tumors beyond breast and gastric cancers. While kinase-domain HER2 mutations are established targets (e.g., neratinib activity in SUMMIT), the therapeutic significance of extracellular domain mutations other than S310F remains unclear. The authors encountered a carcinoma of unknown primary harboring HER2 E401G in the extracellular domain with concurrent HER2 amplification. Prior functional analyses and molecular dynamics suggested E401G leads to HER2 activation via an EGFR-mediated mechanism, similar to S310F. Based on this, the hypothesis was that simultaneous suppression of EGFR and HER2 would be effective. The study aimed to test afatinib (irreversible EGFR/HER2/HER4 TKI) versus lapatinib (reversible EGFR/HER2 TKI with weaker EGFR inhibition) and trastuzumab plus pertuzumab (effective in wild-type HER2 amplification) using patient-derived models (PDX and CTOS) and to contrast results with a wild-type HER2-amplified cell line (H2170).

- Basket trials have expanded targeted therapy application across tumor types with shared driver alterations; MyPathway showed benefit of trastuzumab plus pertuzumab in HER2-amplified tumors across cancers. - SUMMIT trial demonstrated activity of HER2/HER3-directed neratinib in tumors with HER2/HER3 mutations, predominantly kinase-domain alterations. - Extracellular domain HER2 mutations like S310F can signal via EGFR-mediated mechanisms and may respond to dual EGFR/HER2 inhibition; prior PDX data with S310F showed afatinib superiority over trastuzumab-based regimens. - Afatinib is an irreversible ErbB family blocker with potent EGFR inhibition relative to lapatinib; lapatinib binds reversibly and has weaker EGFR potency. - Patient-derived models (PDX, organoids/CTOS) better recapitulate clinical responses versus long-term cell lines, enabling more predictive preclinical testing.

- Models: Established patient-derived xenograft (PDX) and cancer tissue-originated spheroid (CTOS) from a skin biopsy of a patient with carcinoma of unknown primary harboring HER2 E401G and HER2 amplification (IRB 2020-07-R-10; informed consent). PDX generated by subcutaneous implantation into female BRJ (Balb/c Rag-2−/− Jak3−/−) mice with serial in vivo passages (G1–G4). CTOS established from PDX G3 tumor, size-selected (40–70 μm), cultured in StemPro hESC SFM with bFGF and 2-mercaptoethanol, and expanded via at least two in vivo passages to create CTOS lines. - Comparative model: NCI-H2170 lung cancer cell line (wild-type HER2 amplification) used for in vivo xenografts and in vitro assays. - Drugs: Afatinib and lapatinib (LC Laboratories) prepared in 0.5% methylcellulose and administered orally; trastuzumab and pertuzumab administered intraperitoneally in saline. - In vivo efficacy: For PDX, treatment started when tumors reached 150–200 mm3; groups: vehicle, afatinib 20 mg/kg QD 5 days/week, lapatinib 100 mg/kg (and added 150 mg/kg) QD 5 days/week, trastuzumab + pertuzumab (T+P) 30 mg/kg each IP weekly; n=6 per group. For H2170 xenografts, treatment started at 100–200 mm3; groups included vehicle, afatinib 20 mg/kg, lapatinib 150 mg/kg, T+P 30 mg/kg weekly; n=3–4 per group. Tumor volumes measured twice weekly; endpoint day 22 with tumor collection and photography. - Ex vivo/in vitro efficacy: CTOS seeded (~10 CTOS/well) in 96-well plates; drugs added after 24 h; viability (ATP content) measured at day 7 with CellTiter-Glo to derive IC50. H2170 cells seeded 7×10^3 cells/well; drugs added after 24 h; viability at day 3 via Cell Counting Kit-8. - Pathology and profiling: HE and IHC (HER2, CK7, CK20, GATA3) on patient, PDX (G1–G4), and CTOS (G1–G2) tissues. Droplet digital PCR assessed HER2 copy number (HER2 probe dHsaCP1000116; reference RPP30) and HER2 E401G:WT ratio. - Statistics: Data as mean±SD. Wilcoxon rank-sum for two-group comparisons; Kruskal–Wallis with Dunn’s post hoc for three-group comparisons (reported for H2170 xenografts). Dose–response and IC50 using R DRC package. Significance P<0.05. - Ethical compliance: Human studies per Declaration of Helsinki; animal protocols approved (A2021-022-0).

- Model fidelity: PDX (G1–G4) and CTOS (G1–G2) retained morphology, IHC (HER2, CK7/CK20, GATA3), and genetic features of the original tumor, including stable HER2 copy number and predominance of E401G over wild type within the amplified ERBB2. - CTOS drug response: Afatinib showed strongest growth inhibition with IC50 ≈ 0.35 μM; lapatinib was less potent (IC50 ≈ 1.8 μM); trastuzumab+pertuzumab (T+P) showed minimal effect in the CTOS assay. - PDX in vivo efficacy (HER2 E401G + amplification): Afatinib significantly inhibited tumor growth by day 22 versus vehicle. T+P showed a trend toward suppression without statistical significance. Lapatinib (100 or 150 mg/kg) demonstrated limited efficacy compared to afatinib. - Genomic impact: Only afatinib significantly reduced HER2 copy number in PDX tumors at endpoint (ddPCR), whereas lapatinib and T+P did not. - Contrast with wild-type HER2 amplification (H2170): In xenografts, T+P produced the most profound tumor shrinkage, outperforming afatinib and lapatinib; afatinib still significantly reduced tumor size and was superior to lapatinib. In vitro H2170 assays mirrored the trend (T+P relatively weaker in vitro but overall afatinib > lapatinib). - Clinical context: In the index patient, on-entry liquid biopsy showed HER2 CN 2.25 and E401G VAF 6.78%; at end of T+P treatment, CN rose to 2.43 and VAF to 10.9%, paralleling disease progression and limited clinical benefit (best response: stable disease).

Findings support the mechanistic hypothesis that HER2 E401G signals via an EGFR-mediated activation mechanism; thus, dual inhibition of EGFR and HER2 is advantageous. Afatinib’s superiority over lapatinib likely stems from its more potent EGFR inhibition (subnanomolar EGFR IC50) and irreversible binding, providing durable pathway suppression. The poor efficacy of T+P in the E401G-amplified model, despite strong efficacy in wild-type HER2-amplified H2170 xenografts, suggests that extracellular domain mutations like E401G can attenuate the benefit of HER2-only targeting and necessitate EGFR co-inhibition. The significant reduction in HER2 copy number with afatinib implies selective pressure against high-ERBB2-amplified clones under effective dual pathway blockade. Results underscore the value of patient-derived models (PDX, CTOS) to recapitulate clinical behavior and guide precision therapy, especially when cell-line data diverge. Collectively, these data advocate for considering afatinib or similar dual EGFR/HER2 inhibitors in HER2 E401G-amplified cancers and for mechanism-informed treatment selection.

Therapeutic strategies that concurrently suppress EGFR and HER2 signaling are effective against cancers harboring HER2 E401G with ERBB2 amplification and an EGFR-mediated activation mechanism. Afatinib demonstrated superior efficacy over lapatinib and trastuzumab+pertuzumab in patient-derived models, while T+P remained optimal in wild-type HER2-amplified models. These results highlight the importance of elucidating mutation-specific activation mechanisms to implement patient-tailored precision oncology. Future work should validate these findings across additional HER2 extracellular domain mutations and explore optimized combination strategies balancing efficacy and tolerability.

- The study did not directly evaluate other HER2 extracellular domain mutations (e.g., S310F) within the same experimental framework, although prior reports indicate afatinib efficacy for S310F. - Limited tissue availability and necrosis precluded definitive molecular demonstration that EGFR inhibition is the key mediator of afatinib’s activity in E401G-amplified tumors. - Combination therapy of afatinib with trastuzumab/pertuzumab was not tested; while potentially synergistic, clinical tolerability concerns (e.g., diarrhea, skin toxicity) may limit feasibility, as suggested by prior phase I data.