Pre-clinical efficacy of trastuzumab deruxtecan plus sotorasib in KRAS^G12C-mutant non-small cell lung cancer and exploratory HER2 profiling in human tumors

Introduction

Precision treatment of cancers harboring oncogenic KRAS mutation has been revitalized with the discovery of new efficacious small molecules targeting mutant KRAS^G12C. The lead clinical compounds sotorasib (AMG-510) and adagrasib (MRTX849) demonstrated favorable effects with approximately 40% response rates. However, compared to the targeted therapies in EGFR-mutant and ALK-positive non-small cell lung cancer (NSCLC), responses to KRAS^G12C inhibitors (KRAS^G12Ci) tend to be more shallow, and, in most patients, the disease progresses within 5 to 6 months (1-3). Although new KRAS^G12Ci, such as divarasib, olomorasib, and elironrasib, seem to confer higher response rates and longer progression-free survival, resistance to these therapies is inevitable, highlighting the need for newer treatment combinations (4-6).

Antibody-drug conjugates (ADCs), a class of drugs that combine an antibody directed against a specific epitope with a non-specific cytotoxic payload, offer an attractive therapeutic tool for improving responses to targeted therapies, including KRAS^G12Ci (7,8). ADC efficiency is independent of whether the epitope is essential for KRAS^G12Ci resistance, sidestepping the challenge of the complexity of resistance phenotypes. As long as the targeted epitope is expressed by a substantial proportion of tumor cells, the use of an ADC with a payload capable of inducing bystander damage might alleviate the challenge of intra-tumor heterogeneity. Finally, ADCs potentially enable achieving high local concentrations of effective cytotoxic therapies while avoiding their systemic side effects.

Fam-trastuzumab deruxtecan-nxki (T-DXd), an ADC that combines trastuzumab, an antibody that targets human epidermal growth factor receptor 2 (HER2) protein (encoded by the ERBB2 gene), with the topoisomerase 1 inhibitor deruxtecan, is particularly promising. ERBB activity augments KRAS^G12C signaling, promoting tumor progression and reducing the effects of KRAS^G12Ci (9-11). In contrast to trastuzumab, T-DXd responses have been documented in breast cancers (BCs) with low HER2 expression (12). Importantly, similar to the effects observed in EGFR (13) and ALK (14) targeted therapies, KRAS inhibition can lead to a substantial increase in HER2 expression, reflective of both adaptive induction and selective advantage of HER2 overexpressing cells (9). Notably, a combination of T-DXd with ALKi has induced remarkably strong responses in animal models of ALK-positive NSCLC (15). Therefore, T-DXd can be expected to potentiate responses to KRAS^G12Ci.

Based on these premises, we evaluated HER2 protein expression and examined responses to T-DXd-sotorasib combination in a panel of mouse xenograft models of KRAS^G12C-mutant NSCLC. To investigate the translational potential, as an exploratory aim, we examined HER2 protein expression in clinical (pre-treatment and on-treatment) and rapid autopsy (post-treatment) primary and metastatic tumor samples from patients with NSCLC with different driver genomic alteration status. We present this article in accordance with the MDAR and ARRIVE reporting checklists (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1431/rc).

Methods

Xenograft model and treatment

One million cells were suspended in 100 µL of 1:1 RPMI and BME-Type 3 (R&D Systems, # 36-320-1002P) and implanted subcutaneously at two contralateral sites in 4-6-week-old NOD-Scid IL2Rgnull (NSG) mice of both sexes. Tumors measurements were initiated when tumors reached 3–4 mm in diameter. Mice were randomized to control or treatment groups. Sotorasib (50 mg/kg in 0.25% methylcellulose/0.05% Tween-80) was given daily by oral gavages and T-DXd (10 mg/kg) was given intraperitoneally every two weeks. Tumors were measured weekly using calipers, and volumes were calculated assuming spherical shape. Mice were euthanized and tumor tissues were collected. All procedures followed Institutional Animal Care and Use Committee (IACUC)-approved protocol (No. IS00009653R) at H. Lee Moffitt Cancer Center (MCC), in compliance with the institutional guidelines for the care and use of animals. Mice were bred in-house (breeders from Jackson Laboratory) and maintained in Association for the Assessment and Accreditation of Laboratory Animal Care (AAALAC)-accredited, specific pathogen-free facilities with controlled temperature, humidity, 12-hour light/dark cycle.

Study population

The rapid tissue donation (RTD) program at MCC has enabled patients with advanced stage lung cancer to donate tumor tissue samples, as well as body fluids, for research after death (16). As of July 2025, 133 patients were consented for the RTD and autopsies were performed in 67 patients with a mean of 14 hours between death and tissue collection. Samples were processed as formalin-fixed paraffin-embedded (FFPE) and frozen tissue. Inclusion criteria includes: (I) histologically confirmed lung adenocarcinoma or NSCLC, NOS and (II) adequate viable tumor cells (≥10% tumor cells and tumor cell count ≥100). In addition to autopsy samples, all available clinical samples, collected for diagnostic purposes were assessed for adequacy. A total of 161 autopsy samples and 30 clinical samples from 31 patients with NSCLC were analyzed for the present study. Clinical characteristics were extracted from electronic medical records.

MCC Thoracic RTD program was approved by the institutional review board (IRB, Advarra, Columbia, MD, Pro00030829). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Informed consent was obtained from study participants.

HER2 immunohistochemistry (IHC)

In xenograft tumors, IHC was performed on FFPE sections with primary anti-HER2 antibody (Cell Signaling #4290-D8F12, 1:400) and secondary anti-rabbit biotinylated antibody (Vector Labs, BA-1000, 1:100). The tumor proportion score (TPS) was recorded as the percentage of viable tumor cells with partial or complete membrane staining at any intensity.

In human samples, HER2 IHC was performed using the Food and Drug Administration (FDA)-approved in vitro diagnostic (IVD) clone 4B5 as per manufacturer’s protocol (Ventana Medical Systems) (17). The American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) HER2 interpretation guidelines for gastroesophageal adenocarcinoma (GEA; June 2017) (18) is the current clinical standard for HER2 expression assessment in NSCLC. For comparative research purposes, staining results were also interpreted using the BC guidelines (BC; March 2023) (19). Additionally, the research use only (RUO) clone EP1045Y was applied to 34 human tumor samples per manufacturer’s protocol (20). Whole tumor sections containing ≥100 tumor cells were scored as 0, 1+, 2+, or 3+ according to ASCO/CAP guidelines for BC and GEA.

HER2 dual in-situ hybridization

The Ventana HER2/neu Dual ISH DNA Probe Cocktail assay was performed on samples with discordant IHC results or samples with an IHC score of 2+ by both BC and GEA guidelines, using the manufacture’s validated protocol (21). ISH results were documented per ASCO/CAP guidelines (18,19).

See Appendix 1 for methodological details.

Statistical analysis

For the pre-clinical studies, tumor volume plots were generated using Prism 9 (GraphPad Software). Agreement between HER2 scoring guidelines was assessed using McNemar’s Chi-squared test. HER2 expression differences by genomic driver or tissue type were tested using Fisher’s exact or two-sample proportion test. All P values are unadjusted.

Results

T-DXd enhances sotorasib sensitivity in treatment-naive H358 xenograft models of KRASG12C-mutant NSCLC

To assess whether T-DXd can potentiate responses to sotorasib in vivo, we started with a common experimental model of human KRAS^G12C-mutant NSCLC, the H358 cell line. H358 xenografts, subcutaneously implanted into immune-deficient NSG mice, grew rapidly and displayed abundant membranous HER2 expression at baseline (Figure 1A, left). Sotorasib monotherapy induced marked tumor regression (~10-fold tumor reduction), followed by relapse after 10–12 weeks of continuous treatment (Figure 1B, Figure S1A), mirroring patterns seen clinically. Tumors progressing on continuous sotorasib exhibited an additional increase in HER2 expression (Figure 1A), observed as early as 48 hours and maintained during treatment (Figure S1B; Table S1), consistent with our prior in vitro observations (9). Notably, T-DXd monotherapy did not induce regression in therapy-naïve H358 xenografts. In contrast, the sotorasib plus T-DXd combination produced ~1,000-fold volume reduction, with regressions maintained for at least 16 weeks on-treatment (Figure 1B, Figure S2A).

Figure 1 T-DXd enhances sotorasib sensitivity of therapy-naïve and SR H358 xenograft tumors. (A) Representative images of HER2 IHC staining of therapy-naïve and sotorasib-treated tumors. HER2 IHC scoring of whole tumor tissue based on the interpretation guidelines for BC (pink)/GEA (blue) are indicated on the top right corners. Scale bars, 20 μm. (B) Volumetric response dynamics of H358 xenograft tumors to vehicle (control, N=6), 50 mg/kg sotorasib (N=10), 10 mg/kg T-DXd (N=8), and sotorasib/T-DXd combinations (N=10). N denotes individual tumors. Error bars represent mean ± SEM. (C) Volumetric response dynamics of secondary transplants of two independent H358 tumors that have previously progressed on sotorasib to vehicle (control, N=8), sotorasib (N=10), T-DXd (N=6 for SR1, N=8 for SR2), and sotorasib/T-DXd combinations (N=10). Error bars represent mean ± SEM. ****, denotes P<0.0001 of the interaction term of repeated measure ANOVA analysis comparing volumetric responses between sotorasib with sotorasib/T-DXd groups. ANOVA, analysis of variance; BC, breast cancer; GEA, gastroesophageal adenocarcinoma; HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; SEM, standard error of the mean; SR, sotorasib-relapsed; T-DXd, trastuzumab deruxtecan.

T-DXd plus sotorasib combination leads to tumor regression in sotorasib-relapsed (SR) xenograft models

To examine tumors that had relapsed after sotorasib treatment, we continued therapy in the original H358 cohort until regrown tumors reached, on average, pre-therapy volumes (week 21). Adding T-DXd to ongoing sotorasib induced renewed regression that was maintained for an additional 23 weeks of treatment (Figure 1B, Figure S2A).

Two independent models derived from H358 xenografts, established from relapsed tumors and expanded ex vivo (SR1 and SR2), retained high baseline HER2 expression that further increased with sotorasib exposure (Figure S2B). In these models, sotorasib monotherapy produced partial regressions that were less profound and less durable than in therapy-naïve H358 xenografts (Figure 1C, Figure S2C). T-DXd monotherapy did not induce regressions, whereas the sotorasib-T-DXd combination restored deep, durable regression comparable to those in therapy-naïve xenografts (Figure 1C, Figure S2C).

Activity of T-DXd plus sotorasib across additional KRASG12C-mutant xenograft models

Given the substantial inter-tumor variability of KRAS^G12C-mutant NSCLC, we extended our analyses to 8 additional xenograft models (Figure 2A). Consistent with prior reports from in-vitro studies (22), we observed variability in responses to sotorasib across individual xenograft models. While none of these additional models matched strong sotorasib sensitivity observed in H358 xenografts, 5/8 displayed transient stabilization of the tumor volume or modest regression, with progression occurring within 1–5 weeks (Figure S3A). In 6/8 models, T-DXd monotherapy inhibited tumor growth but failed to induce regression. In contrast, sotorasib-T-DXd combination induced regression in 7/8 of the additional KRAS^G12C xenografts (Figure S3B,S3C). Notably, sotorasib- T-DXd did not induce overt signs of toxicity and weight loss (Figure S3D).

Figure 2 Relationship between HER2 expression levels and sotorasib-T-DXd combination sensitivity. (A) Representative images of HER2 IHC of therapy-naïve and sotorasib-treated tumors from the indicated xenograft models of KRAS^G12C-mutant NSCLC. HER2 IHC scoring of whole tumor tissue based on the interpretation guidelines for BC (pink)/GEA (blue) are indicated on the top right. Lu65 and H2122 are highlighted with an asterisk (*) in the figure. Scale bars, 20 μm. (B,C) Maximal magnitude of tumor regression in response to the indicated therapies as a function of the IHC-based scoring (BC and GEA) of therapy-naïve (B) and sotorasib-treated (C) tumors from the nine KRAS^G12C-mutant xenograft models. IHC scores using both the BC and GEA interpretation guidelines were concordant in therapy-naïve samples, but discrepancies between the two interpretation guidelines were observed in sotorasib-treated tumors. BC, breast cancer; FC, fold change; GEA, gastroesophageal adenocarcinoma; HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; NSCLC, non-small cell lung cancer; T-DXd, trastuzumab deruxtecan.

None of the KRAS^G12C-mutant cell line-derived xenograft models used in this study harbor known activating ERBB2 (HER2) kinase domain mutations. Thus, the observed activity of T-DXd in combination reflects HER2 protein expression rather than HER2 mutational status.

Exploratory correlation with HER2 expression in KRASG12C-mutant xenograft models

Given that T-DXd sensitivity requires HER2 expression and that clinical guidelines for IHC-based assessment of HER2 expression in NSCLC have not yet been developed (23), we compared HER2 IHC analysis [using both GEA (18) and BC (19) guidelines] with treatment outcomes. Visual review showed increased HER2 staining after sotorasib in some tumors (Figure 2A), though both interpretation guidelines often classified pre-and post-sotorasib samples as IHC 3+, potentially obscuring changes. While for therapy-naïve tumors, GEA and BC interpretation guidelines yielded identical scores, some divergence was noted in post-sotorasib tumors (Figure 2A, asterisks).

To assess the correlation between HER2 expression and therapeutic sensitivities, we plotted the maximal magnitude of tumor regression (group average fold change in tumor volume) as a function of HER2 IHC scores. No clear correlation was observed for sotorasib or T-DXd monotherapies (Figure 2B). The sotorasib-T-DXd combination showed a trend toward greater regressions in tumors with higher HER2 expression post-sotorasib, though these analyses were not powered to detect a difference due to limited sample size (Figure 2C). To account for the intra-tumor variability in the HER2 expression, we performed TPS, which is commonly used to quantify programmed death ligand-1 (PD-L1) expression in NSCLC and to identify patients eligible for immune checkpoint inhibitor therapy (24). TPS increased post-sotorasib in most models (Table S1). Notably, T-DXd monotherapy lead to reduced HER2 expression, consistent with the therapeutic escape via target downregulation (Figure S4). These exploratory results suggest HER2 expression may contribute to the combination therapy efficacy while highlighting the limitations of the available HER2 IHC interpretation guidelines in NSCLC.

Patient cohort characteristics

The robust sotorasib-T-DXd combination activity in KRAS^G12C-mutant NSCLC xenograft models provided rationale for a first-in-human phase I/II clinical trial evaluating this combination in patients with previously treated KRAS^G12C-mutant NSCLC (NCT07012031). To inform biomarker strategy, we characterized HER2 IHC expression in tumors from 31 patients (Table 1). Median age was 62 years; 97% had adenocarcinoma, 55% had a KRAS mutation, and no HER2 mutation or amplification were detected. Samples included clinical biopsies (N=30) and autopsy samples (N=161).

HER2 expression in NSCLC using ASCO/CAP interpretation guidelines developed for BC and GEA

Across 191 tumor samples, HER2 IHC scores (0, 1+, 2+, and 3+) were distributed as 70%, 10%, 12%, 7.9% (BC) and 68%, 2%, 20%, 10% (GEA) (Table 2). HER2 2+/3+ expression was identified in 19.9% and 30% of samples by BC and GEA guidelines, respectively. Concordance between BC and GEA guidelines was 87%, with 13% discordance (Table 2, Figure S5). Nuclear or cytoplasmic HER2 was noted in a minority of tumors.

HER 2+/3+ expression was numerically higher in KRAS-mutant tumors compared with those harboring other driver genomic alterations (ALK, EGFR, MET, NRAS, and ROS1), though not statistically significant (Table 3). There were no differences in HER2 expression between KRAS^G12C-mutant and non-G12C-mutant NSCLC (Table S2).

Variability in HER2 expression in NSCLC patient’s samples

HER2 IHC scoring in patient samples demonstrated variability by both specimen type and assay conditions. HER2 2+/3+ expression was more frequent in clinical samples compared with autopsy samples when scored by the BC guideline (33% vs. 17%, P=0.04), while the difference was not significant with the GEA guideline. Furthermore, HER2 antibody clone selection influenced scoring outcomes: the RUO clone identified markedly higher rates of HER2 2+/3+ positivity than the IVD clone (58–69% vs. 7.7–12%, P<0.001) (Figure S6, Table S3).

Further analysis—including clinical vs. autopsy comparisons, comparison of HER2 IHC scores in pre-treatment (clinical) vs. post-treatment (autopsy) samples from the same tumor, antibody clone assessment (IVD vs. RUO) HER2, and ISH results—are provided in Appendix 1, Tables S3,S4.

Discussion

In this study, we demonstrate that combining sotorasib with T-DXd produced robust and durable regressions across multiple KRAS^G12C-mutant NSCLC xenograft models. While sotorasib monotherapy initially induced tumor regressions in the H358 model, progression inevitably occurred during treatment. The addition of T-DXd not only deepened responses but also prolonged tumor control far beyond either agent alone. Importantly, the combination remained active in derivative models established from tumors that had relapsed during continuous sotorasib exposure, suggesting preserved therapeutic vulnerability in this setting.

Mechanistically, xenografts progressing on sotorasib displayed increased HER2 expression, detectable as early as 48 hours after treatment and sustained during therapy. Although T-DXd was ineffective as monotherapy in these models, its combination with sotorasib restored deep and durable regressions. It is important to note that lack of tumor regression in our xenograft models does not necessarily contradict clinical experience. In DESTINY-Lung01, although objective responses were observed in a substantial proportion of patients with HER2-overexpressing NSCLC, not all HER2 IHC 3+ tumors demonstrated treatment response. A meaningful subset of patients achieved stable disease, and a small proportion experienced progressive disease despite high tumoral HER2 expression. Consistent with this, T-DXd monotherapy in our models induced modest regression and growth stabilization in the majority of models. Notably, T-DXd treated tumors downregulated HER2 expression (Figure S4), consistent with therapeutic escape through target downregulation. Given the HER2 upregulation under sotorasib and the importance of HER2 signaling for sotorasib persistence, the apparent synergy between sotorasib and T-DXd treatment might reflect a scenario of a therapeutic double bind (25), which limits therapeutic escape to both agents.

Our findings underscore that HER2 IHC 3+ status alone does not uniformly predict depth of response and that additional biological factors are likely to influence therapeutic sensitivity. These observations support a model in which sotorasib-mediated HER2 modulation may sensitize tumors to HER2-directed ADC therapy. However, these findings are exploratory and do not establish HER2 upregulation as the sole or necessary mechanism underlying combination efficacy.

While the mechanistic underpinnings of the observed efficacy of the sotorasib-T-DXd combination, as well as the biological drivers of model-to-model variability remain to be duly defined, our preclinical data provided the rationale for a first-in-human Phase I/II clinical trial (NCT07012031). Observations supporting the effectiveness of KRAS-MAPK pathway inhibitors in combination with HER2-directed antibody–drug conjugates in pancreatic cancer (26) further strengthen the rationale for this clinical trial. This investigator-initiated trial will evaluate the safety and efficacy of sotorasib plus T-DXd in patients with advanced, previously treated KRAS^G12C-mutant NSCLC. This represents a critical step toward translating our preclinical observations into a potential new treatment option for this challenging disease.

Several alternative combination strategies are currently being explored in KRAS^G12C-mutant NSCLC, including KRAS^G12Ci combined with immunotherapy+/− chemotherapy or SHP2 inhibitors. While KRAS^G12Ci-immunotherapy combinations are biologically appealing, many patients with advanced NSCLC receive immune checkpoint inhibitors in the first-line setting, limiting the applicability of this approach in later lines of therapy. The ongoing phase I/II study of sotorasib plus T-DXd (NCT07012031) is specifically designed for patients previously exposed to both KRAS^G12C inhibition, chemotherapy, and immunotherapy, positioning this regimen as a potential second-line and beyond therapeutic strategy if clinical benefit is confirmed. SHP2-based combinations represent another rational approach to overcoming adaptive signaling; however, emerging clinical data suggest that toxicity may limit feasibility in some regimens. In contrast, the mostly non-overlapping toxicity profiles of sotorasib and T-DXd support clinical exploration of this combination. Importantly, the current trial incorporates exploratory analyses to evaluate whether baseline HER2 IHC expression should inform patient selection, and the dose-escalation phase is expected to provide preliminary insight into the predictive value of HER2 expression in this context.

Clinical translation of this combination requires careful consideration of toxicity profiles. Sotorasib is associated primarily with hepatotoxicity, including transaminase elevations, whereas T-DXd carries a risk of interstitial lung disease (ILD)/pneumonitis. These toxicities are mechanistically distinct and not expected to be overlapping. In the ongoing phase I/II trial (NCT07012031), a dose-escalation design has been implemented to formally define the safety and tolerability of the combination. Liver function tests are prospectively monitored given the known hepatic effects of sotorasib, and patients undergo close clinical assessment for pulmonary symptoms with predefined ILD management algorithms. These safeguards are intended to ensure safe clinical evaluation while maintaining the potential therapeutic benefit observed in preclinical models.

Our broader xenograft screen further supports generalizability, as most KRAS^G12C models benefited from the combination therapy despite heterogeneous baseline responses to either agent. Notably, HER2 IHC scoring using breast and GEA interpretation guidelines did not reliably distinguish responders from non-responders in NSCLC xenografts, underscoring the limitations of applying current ASCO/CAP HER2 IHC interpretation guidelines to NSCLC. The absence of a strong correlation between categorical HER2 IHC scores and magnitude of tumor regression suggests that current scoring systems may lack sufficient resolution to serve as predictive biomarkers in this context. HER2 IHC interpretation guidelines were developed for other tumor types and rely on semi-quantitative categorical thresholds that may not capture biologically meaningful variability in NSCLC. In our exploratory analyses using TPS, we observed substantial intratumoral heterogeneity even among tumors classified as IHC 2+ or 3+, indicating that HER2 expression exists along a continuum rather than discrete categories. While HER2 expression is required for T-DXd binding and internalization, our data does not establish that HER2 upregulation alone is sufficient to determine depth of response to the combination. It is therefore possible that HER2 functions as a permissive factor rather than a sole driver of efficacy. The ongoing phase I/II trial incorporates prospective exploratory analyses of baseline HER2 expression to more definitively evaluate its predictive value in patients with KRAS^G12C-mutant NSCLC.

Given the expected dependence of responses to sotorasib-T-DXd combination on HER2 expression, we also explored HER2 expression in a patient cohort with NSCLC. HER2 2+/3+ expression was identified in 19.9% and 30% of samples by BC and GEA guidelines, respectively. In comparison, the prevalence of HER2 2+/3+ expression in NSCLC has previously been reported as approximately 12% using the BC guideline with same clone (4B5) (27). Approximately 60% of the patients with KRAS-mutant NSCLC in this study had specimens that displayed HER2 2+/3+ expression using either guideline, a prevalence higher than unselected NSCLC groups. While concordance between breast and gastroesophageal HER2 IHC scoring systems was generally high, discordance was observed in a minority of cases, reflecting the challenges of adapting these frameworks to NSCLC. Although HER2 expression appeared numerically enriched in KRAS-mutant tumors compared with other oncogenic subsets, these findings did not reach statistical significance and require validation in larger cohorts.

We also observed notable differences in HER2 expression between clinical and autopsy samples. Clinical samples demonstrated a higher prevalence of HER2 expression, raising the possibility that HER2 expression may decline over time, during treatment, or with disease progression. Although prior studies suggest rapid autopsy protocols preserve protein antigenicity (16), reduced IHC positivity could also result from postmortem processes such as autolysis rather than true tumor biology (28).

Another source of variability was the choice of HER2 IHC antibody clones. The RUO clone EP1045Y identified a markedly higher proportion of HER2-positive tumors than the FDA approved 4B5 IVD clone, with discrepancies evident using both the BC and GEA interpretation guidelines. These findings highlight the need for assay harmonization in NSCLC, as differences in antibody selection and scoring criteria could directly impact patient classification and, ultimately, treatment decisions.

T-DXd is currently approved in NSCLC for both tumors harboring activating HER2 mutations and for tumors with HER2 overexpression. HER2 mutation is not mechanistically required for ADC binding and internalization, which depend on HER2 protein expression at the cell surface. In our preclinical models, HER2 modulation occurred at the level of protein expression without evidence of ERBB2 mutational events, consistent with prior reports that adaptive ERBB signaling following KRAS inhibition is driven by non-mutational mechanisms.

Our study has several limitations. We focused primarily on xenograft efficacy and HER2 expression, without addressing other potential mechanisms of variable response. The IHC analysis was exploratory, cross-sectional, and limited in sample size. Importantly, while HER2 modulation by sotorasib was observed preclinically, we cannot yet determine whether the same phenomenon occurs in patients.

Conclusions

Our findings provide a compelling rationale for clinical evaluation of sotorasib in combination with T-DXd in KRAS^G12C-mutant NSCLC, a strategy now being prospectively tested. The combination induced deep and durable tumor regressions across both sotorasib-naïve and SR xenograft models, supporting its potential to overcome the limited depth and durability of response observed with KRAS^G12C inhibitor monotherapy. Exploratory analyses of patient samples suggest that HER2 expression is sufficiently prevalent to support biomarker-driven investigation; however, the inconsistency of current HER2 IHC interpretation frameworks in NSCLC underscores a critical need for assay optimization and standardization. Taken together, these data define a biologically grounded and clinically actionable combination strategy, while highlighting the importance of refining HER2-directed biomarkers to enable precise patient selection and maximize therapeutic impact in KRAS^G12C-mutant NSCLC.

Acknowledgments

We would like to thank Angela Todica for helping us procure T-DXd, Pragya Kumar for her technical assistance with sample collection, Amer A. Beg for securing grants and funding, and Alberto A. Chiappori, Andreas N. Saltos, Benjamin C. Creelan, Charles C. Williams, George Simon, Jhanelle E. Gray, Michael R. Shafique, Scott J. Antonia, and Tawee Tanvetyanon for their efforts in recruiting patients, as well as Rob Macaulay for helping with autopsies for the RTD program. We would like to extend our heartfelt appreciation to the patients and their loved ones who have graciously provided consent for participation in the RTD. During the preparation of this work the authors used ChatGPT to improve readability and language. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Footnote

Reporting Checklist: The authors have completed the MDAR and ARRIVE reporting checklists. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1431/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1431/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1431/prf

Funding: The Rapid Tissue Donation (RTD) protocol was funded by Bristol Myers Squibb, and both RTD protocol and this study were funded by the Moffitt Lung Cancer Center of Excellence. The Tissue Core, and Biostatistics and Bioinformatics Shared Resource are funded in part by the NCI Cancer Center Support grant, which confers Moffitt’s status as an NCI-designated Comprehensive Cancer Center (NCI P30-CA076292). Preclinical studies were supported by the Moffitt Lung Cancer Center of Excellence (NIH UO1 CA280829 and NIH U54CA274507).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-1-1431/coif). E.B.H. receives research support from Revolution Medicines, reports advisory board work with Amgen, Janssen and Revolution Medicines and consulting work for Ellipses, Kanaph Therapeutics and ORI Capital II. B.P. has received research support to the institution from Merck, Bristol Myers Squibb, Orbit Genomics, and the Bristol Myers Squibb Foundation/the Robert A. Winn Diversity in Clinical Trials Awards Program, speaker honoraria from AstraZeneca, Amgen, Bayer, Merck, Foundation Medicine, Gilead, Regeneron, Oncohost, and has done consulting/advisory board work with AstraZeneca, AbbVie, Bayer, Bristol Myers Squibb, Boehringer-Ingelheim, Catalyst, Caris, Gilead, Guardant Health, Foundation Medicine, Illumina, J&J, Lilly, Regeneron, Merus, Oncohost, and ThermoFisher. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures followed Institutional Animal Care and Use Committee (IACUC)-approved protocol (No. IS00009653R) at H. Lee Moffitt Cancer Center, in compliance with the institutional guidelines for the care and use of animals. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Informed consent was obtained from study participants. Tissue samples were procured in line with WHO Guiding Principles on Human Cell, Tissue and Organ Transplantation. Moffitt Cancer Center (MCC) Thoracic Rapid Tissue Donation (RTD) program was approved by the institutional review board (IRB, Advarra, Columbia, MD, Pro00030829).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Skoulidis F, Li BT, Dy GK, et al. Sotorasib for Lung Cancers with KRAS p.G12C Mutation. N Engl J Med 2021;384:2371-81. [Crossref] [PubMed]
2. de Langen AJ, Johnson ML, Mazieres J, et al. Sotorasib versus docetaxel for previously treated non-small-cell lung cancer with KRAS(G12C) mutation: a randomised, open-label, phase 3 trial. Lancet 2023;401:733-46. [Crossref] [PubMed]
3. Jänne PA, Riely GJ, Gadgeel SM, et al. Adagrasib in Non-Small-Cell Lung Cancer Harboring a KRAS(G12C) Mutation. N Engl J Med 2022;387:120-31. [Crossref] [PubMed]
4. Sacher A, LoRusso P, Patel MR, et al. Single-Agent Divarasib (GDC-6036) in Solid Tumors with a KRAS G12C Mutation. N Engl J Med 2023;389:710-21. [Crossref] [PubMed]
5. Burns TF, Dragnev KH, Fujiwara Y, et al. Efficacy and safety of olomorasib (LY3537982), a second-generation KRAS G12C inhibitor (G12Ci), in combination with pembrolizumab in patients with KRAS G12C-mutant advanced NSCLC. J Clin Oncol 2024;42:8510. [Crossref]
6. Jänne PA, Bigot F, Papadopoulos K, et al. Abstract PR014: Preliminary safety and anti-tumor activity of RMC-6291, a first-in-class, tri-complex KRASG12C(ON) inhibitor, in patients with or without prior KRASG12C(OFF) inhibitor treatment. Mol Cancer Ther 2023;22:PR014. [Crossref]
7. Dumontet C, Reichert JM, Senter PD, et al. Antibody-drug conjugates come of age in oncology. Nat Rev Drug Discov 2023;22:641-61. [Crossref] [PubMed]
8. Chau CH, Steeg PS, Figg WD. Antibody-drug conjugates for cancer. Lancet 2019;394:793-804. [Crossref] [PubMed]
9. Solanki HS, Welsh EA, Fang B, et al. Cell Type-specific Adaptive Signaling Responses to KRAS(G12C) Inhibition. Clin Cancer Res 2021;27:2533-48. [Crossref] [PubMed]
10. Ho CSL, Tüns AI, Schildhaus HU, et al. HER2 mediates clinical resistance to the KRAS(G12C) inhibitor sotorasib, which is overcome by co-targeting SHP2. Eur J Cancer 2021;159:16-23. [Crossref] [PubMed]
11. Kruspig B, Monteverde T, Neidler S, et al. The ERBB network facilitates KRAS-driven lung tumorigenesis. Sci Transl Med 2018;10:eaao2565. [Crossref] [PubMed]
12. Modi S, Jacot W, Yamashita T, et al. Trastuzumab Deruxtecan in Previously Treated HER2-Low Advanced Breast Cancer. N Engl J Med 2022;387:9-20. [Crossref] [PubMed]
13. Takezawa K, Pirazzoli V, Arcila ME, et al. HER2 amplification: a potential mechanism of acquired resistance to EGFR inhibition in EGFR-mutant lung cancers that lack the second-site EGFRT790M mutation. Cancer Discov 2012;2:922-33. [Crossref] [PubMed]
14. Minari R, Gnetti L, Lagrasta CA, et al. Emergence of a HER2-amplified clone during disease progression in an ALK-rearranged NSCLC patient treated with ALK-inhibitors: a case report. Transl Lung Cancer Res 2020;9:787-92. [Crossref] [PubMed]
15. DesaiBMitiTPrabhakaranSPeristromal niches protect lung cancers from targeted therapies through a combined effect of multiple molecular mediators. Preprint. bioRxiv. 2024;2024.04.24.590626[Preprint]. 2024. doi:10.1101/2024.04.24.590626
16. Boyle TA, Quinn GP, Schabath MB, et al. A community-based lung cancer rapid tissue donation protocol provides high-quality drug-resistant specimens for proteogenomic analyses. Cancer Med 2020;9:225-37. [Crossref] [PubMed]
17. PATHWAY anti-HER2/neu (4B5) Rabbit Monoclonal Primary Antibody. 2024. Available online: https://diagnostics.roche.com/us/en/products/lab/her-2-neu-4b5-pathway-rtd001084.html
18. Bartley AN, Washington MK, Colasacco C, et al. HER2 Testing and Clinical Decision Making in Gastroesophageal Adenocarcinoma: Guideline From the College of American Pathologists, American Society for Clinical Pathology, and the American Society of Clinical Oncology. J Clin Oncol 2017;35:446-64. [Crossref] [PubMed]
19. Wolff AC, Somerfield MR, Dowsett M, et al. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: ASCO-College of American Pathologists Guideline Update. J Clin Oncol 2023;41:3867-72. [Crossref] [PubMed]
20. Anti-ErbB2/HER2 antibody [EP1045Y]. Available online: https://www.abcam.com/en-us/products/primary-antibodies/erbb2-her2-antibody-ep1045y-ab134182
21. VENTANA HER2 Dual ISH DNA Probe Cocktail. Available online: https://diagnostics.roche.com/us/en/products/lab/her2-dual-ish-dna-probe-cocktail-ventana-rtd001251.html
22. Kato R, Solanki HS, Ozakinci H, et al. In Situ RAS:RAF Binding Correlates with Response to KRASG12C Inhibitors in KRASG12C-Mutant Non-Small Cell Lung Cancer. Clin Cancer Res 2025;31:1150-62. [Crossref] [PubMed]
23. Ren S, Wang J, Ying J, et al. Consensus for HER2 alterations testing in non-small-cell lung cancer. ESMO Open 2022;7:100395. [Crossref] [PubMed]
24. NCCN NSCLC Guidelines, Version 3.2025. Available online: https://www.nccn.org/guidelines/guidelines-detail?category=1&id=1450
25. Gatenby RA, Brown J, Vincent T. Lessons from applied ecology: cancer control using an evolutionary double bind. Cancer Res 2009;69:7499-502. [Crossref] [PubMed]
26. Bulle A, Liu P, Seehra K, et al. Combined KRAS-MAPK pathway inhibitors and HER2-directed drug conjugate is efficacious in pancreatic cancer. Nat Commun 2024;15:2503. [Crossref] [PubMed]
27. Abu Al Karsaneh O, Al Anber A. Prevalence and clinicopathological associations of HER2 expression in non-small cell lung cancer: a retrospective study in Jordanian patients. Diagn Pathol 2023;18:75. [Crossref] [PubMed]
28. Shedge R, Krishan K, Warrier V, et al. Postmortem Changes. Treasure Island (FL): StatPearls Publishing; 2023.