Uveal melanoma originates in the uvea, a pigmented tissue layer situated between the retina and the sclera, or the eye’s white outer layer. Its clinical course is often aggressive; nearly half of patients develop metastases, predominantly in the liver, where therapeutic options are distressingly limited. The inability of current mouse models to recapitulate this disease’s natural history has impeded translational progress. By genetically engineering mice to harbor the same sequential mutations characteristic of human uveal melanoma—starting with the activation of the GNAQ oncogene, followed by deletion of the tumor suppressor gene BAP1, and culminating in MYC oncogene amplification—researchers have created a system that faithfully replicates tumor initiation, progression, and phenotypic diversity.

This model represents an intricate approach to cancer modeling, activating a GNAQ mutation that alone induces benign ocular lesions akin to nevi in patients. The subsequent loss of BAP1 triggers malignant transformation, and amplification of MYC correlates with heightened tumor aggression and histopathological features resembling those seen in lethal human cases. This stepwise genetic manipulation underscores the multi-hit hypothesis of oncogenesis, elucidating how cumulative alterations drive malignancy’s advancement while preserving physiological relevance by maintaining an intact immune system.

Intriguingly, the study elucidates the phenotypic plasticity of uveal melanoma cells. Cancer cells within the tumors do not constitute a homogenous population; instead, subpopulations exhibit distinct states. Some retain characteristics similar to normal melanocytes, while others adopt aggressive phenotypes associated with poor clinical outcomes. This cellular heterogeneity likely contributes to the tumor’s notorious resilience and capacity for metastasis. The model facilitates in-depth dissection of how tumors shift cellular states dynamically, possibly in response to environmental pressures or therapeutic interventions, mirroring phenomena previously described in cutaneous melanoma.

A particularly compelling aspect of the research involves the immune microenvironment. Both in this mouse model and human tumors, immune cells infiltrate the tumor but remain dysfunctional, effectively stymied by the cancer’s immunosuppressive tactics. Such immune evasion tactics help explain why conventional immunotherapy, so successful in other melanoma types, remains largely ineffective for uveal melanoma. By reproducing this immune landscape, the model opens avenues for developing tailored immunotherapies, designed to overcome the unique barriers found in ocular tumors.

Furthermore, researchers identified molecular biomarkers linked to aggressive tumor phenotypes within the model. These biomarkers offer potential for refining prognostication and personalizing treatment protocols. Existing clinical tools inadequately predict metastatic risk, and these newly discovered biomarkers, grounded in a replicable in vivo system, promise to enhance risk stratification and catalyze biomarker-driven clinical trials. This could herald a new era of precision medicine in eye cancer treatment.

The model’s ability to mimic tumor spread to the liver—albeit initially without extensive metastatic outgrowth—makes it a valuable tool for probing the mechanisms underlying organ tropism. Understanding why uveal melanoma cells preferentially colonize the liver, while sparing other organs, remains a significant scientific puzzle. Researchers hypothesize that disseminated cells undergo state transitions that enable migration and colonization, subsequently reverting to a proliferative state to establish secondary tumors. This model provides an experimental venue to test these hypotheses systematically, potentially revealing interventions to disrupt metastatic colonization or dormancy escape.

Beyond the insights into tumor biology, this immune-competent and genetically engineered mouse model equips the scientific community with a tool to evaluate novel therapeutic regimens in a physiologically relevant context. It supports studies that investigate immune checkpoint inhibitors, adoptive cell therapies, and combination treatments tailored to the unique genetic and immunological features of uveal melanoma. By enabling preclinical screening of immunotherapies before human trials, this model may accelerate the advent of effective treatments for a cancer that currently offers a grim prognosis.

The stepwise approach taken to model uveal melanoma genetics aligns with best practices established in other cancer research fields. Incorporating multiple patient-relevant mutations and maintaining an intact functional immune system enhances the model’s clinical relevance. It underscores a paradigm where preclinical studies leverage genetically engineered mouse models that recapitulate the heterogeneity and complexity of human cancers to optimize translational potential. This approach is likely to inspire similar strategies across diverse malignancies requiring nuanced modeling.

A crucial advancement made by the scientists is their ability to restrict the effects of oncogenic mutations in spatially and temporally controlled manners, ensuring that early benign lesions form similarly to human nevi before malignant progression. This refinement contrasts with earlier models where immediate tumor formation skewed interpretations and failed to reproduce disease kinetics. This nuanced control allows researchers to dissect initiation, progression, dormancy, and metastasis phases with unprecedented granularity.

The identification of phenotypic plasticity within uveal melanoma cells invites further exploration of epigenetic mechanisms and signaling pathways that regulate state transitions. Understanding these processes could reveal vulnerabilities exploitable for therapeutic intervention, such as targeting state-switching machinery to prevent metastasis or therapy resistance. The model offers a robust platform for interrogating these dynamic cancer cell behaviors, advancing efforts to counteract tumor adaptability.

In conclusion, the development of this multi-step, immune-competent genetically engineered mouse model represents a landmark accomplishment in ocular melanoma research. Its capacity to recapitulate tumor genetics, cellular heterogeneity, immune interactions, and metastatic behavior provides a transformative tool to unravel the baffling biology of uveal melanoma. By enabling rigorous preclinical testing of targeted and immunotherapeutic approaches, this model holds promise for accelerating the discovery of life-saving treatments for patients afflicted by this devastating eye cancer.

Subject of Research: Animals

Article Title: A Multi-Step Immune-Competent Genetically Engineered Mouse Model Reveals Phenotypic Plasticity in Uveal Melanoma

News Publication Date: 10-Feb-2026

Web References:

• Moffitt Cancer Center: http://moffitt.org/
• Uveal Melanoma Information: https://www.moffitt.org/cancers/melanoma/diagnosis/types/ocular-melanoma/
• Research Article in Cancer Research: https://aacrjournals.org/cancerres/article/doi/10.1158/0008-5472.CAN-25-2684/774239/A-Multi-Step-Immune-Competent-Genetically

References:
Karreth, F., et al. (2026). A Multi-Step Immune-Competent Genetically Engineered Mouse Model Reveals Phenotypic Plasticity in Uveal Melanoma. Cancer Research. DOI: 10.1158/0008-5472.CAN-25-2684

Keywords: Eye cancers, uveal melanoma, mouse model, immune microenvironment, cancer genetics, GNAQ mutation, BAP1 deletion, MYC amplification, phenotypic plasticity, metastasis, immunotherapy, tumor biomarkers

Small-cell lung cancer is a formidable neuroendocrine tumor with distinct molecular subtypes defined by differential expression of transcription factors such as ASCL1 and POU2F3. While MYC amplification has been long associated with poor prognosis and aggressive tumor behavior, the relationship between MYC, PTEN loss, and the resulting phenotypic heterogeneity in SCLC has remained obscure. This latest study bridges that gap by revealing that PTEN loss preferentially drives the emergence of POU2F3-high tumors, a shift marked by upregulated MYC and depletion of ASCL1 expression.

The researchers began by meticulously analyzing a cohort of 112 human SCLC tumors, stratifying them based on POU2F3 expression levels. They discovered that PTEN deletion was significantly more prevalent in POU2F3-high tumors, occurring in 63% of cases compared with only 27% in POU2F3-low tumors. This statistically significant finding (P < 0.009) suggests a tight genetic linkage between PTEN loss and the SCLC-P subtype, which features tuft cell-like characteristics. Importantly, this correlation provides a genetic basis to previously observed phenotypic differences within SCLC subgroups, emphasizing the role of PTEN in the tumor lineage landscape.

To experimentally validate these observations, the team utilized CRISPR-Cas9 gene editing to knockout PTEN in basal organoids derived from RPM and RPMA mouse models. RPM tumors are characterized by the expression of ASCL1, whereas RPMA tumors lack ASCL1 and express YAP1, resembling human SCLC-P. PTEN loss in these models led to accelerated tumor growth, confirming PTEN’s role as a potent tumor suppressor in lung cancer. Intriguingly, while YAP1 and ASCL1 expression remained stable following PTEN deletion, POU2F3 expression markedly increased in both organoid types, underscoring PTEN loss as a driver of the tuft cell-like SCLC phenotype.

Immunohistochemistry analyses of RPMA tumors with PTEN deletion revealed increased POU2F3 expression near-uniformly across tumor cells. The elevation of POU2F3 correlated closely with phospho-AKT levels—signifying activated PI3K/AKT signaling pathways downstream of PTEN loss—and inversely correlated with NEUROD1, another neuroendocrine lineage marker. This inverse relationship indicates that PTEN loss not only enhances SCLC-P features but seemingly suppresses alternate neuroendocrine fates, specifically the SCLC-N subtype characterized by NEUROD1 expression.

Beyond the molecular phenotype, PTEN-deleted RPMA tumors exhibited striking histological heterogeneity, comprising regions of adenocarcinoma, adenosquamous carcinoma, and squamous cell carcinoma interspersed within predominantly small-cell histology. Notably, these non-small cell lung cancer (NSCLC) regions were enriched for basal cell markers KRT5 and P63, suggesting a lineage drift influenced by PTEN loss and MYC activity. This phenotypic plasticity mirrors clinical observations where SCLC-P can be found adjacent to squamous cell carcinoma in combined SCLC, implying possible transitional states between these histologies.

The emergence of squamous-like and tuft-like features within the same tumors raises compelling questions about the role of ASCL1 deficiency in facilitating divergent lineage choices under MYC and AKT signaling pressure. Since squamous cell carcinomas often originate from basal cells and exhibit active MYC and PI3K/AKT pathways, the results suggest that ASCL1 status might gate the cellular trajectory toward either neuroendocrine tuft cells or squamous epithelial differentiation. This finding not only enriches the biological understanding of SCLC heterogeneity but also opens avenues for lineage-targeted therapies.

Further supporting these conclusions, the authors employed genetically engineered mouse models (GEMMs) and demonstrated that induction of lung cancer through K5-Cre recombinase in the presence of PTEN loss favored POU2F3-high tumor development. Tumors from these mice displayed a robust correlation between MYC and POU2F3 expression, reinforcing the cooperative effect of MYC amplification and PTEN deficiency in driving SCLC-P fate. Conversely, tumors with lower MYC levels expressed less POU2F3, emphasizing the dose-dependent nature of the genetic interplay.

Immunohistochemistry for subtype markers in K5-Cre-induced tumors further revealed that high-MYC regions aligned with POU2F3 positivity, whereas low-MYC regions were devoid of this expression. This regional heterogeneity within tumors highlights the spatial dynamics of transcription factor expression and lineage commitment during tumor progression. It also suggests that therapeutic strategies modulating MYC or its downstream effectors could adjust tumor cell fate and sensitivity to treatment.

From a broader perspective, this study exemplifies how precise genomic edits in defined cell populations can clarify the contribution of genetic drivers to tumor lineage choice and plasticity. The use of basal cell-derived organoids and animal models allowed the authors to dissect the cell-intrinsic effects of genetic alterations, minimizing confounding influences such as tumor microenvironment variability. This approach advances the field toward more sophisticated models of tumor heterogeneity that better recapitulate human disease.

Clinically, the link between PTEN loss and the SCLC-P subtype carries profound implications. The SCLC-P subtype tends to resist traditional neuroendocrine-targeted therapies, and its connection to hyperactivated PI3K/AKT signaling suggests that targeting this pathway might yield therapeutic benefits. Moreover, the coexistence of tuft-like and squamous-like tumors within single lesions calls for reassessment of diagnostic criteria and therapeutic regimens, advocating for personalized treatments based on detailed molecular profiling.

In sum, this comprehensive investigation sheds light on the molecular underpinnings of SCLC subtype specification, revealing that the intersection of PTEN loss, MYC gain, and cell of origin decisively sculpts tumor phenotype and behavior. The consequent model of lineage plasticity not only advances fundamental cancer biology but also equips clinicians with new conceptual tools to tackle one of the deadliest lung cancers with tailored strategies.

As research continues to unravel the complexities of lung cancer subtypes, studies like this illuminate the path toward precision oncology, where understanding the genetic and cellular context of tumors enables more effective and less toxic therapies. The elucidation of PTEN and MYC’s convergent roles in defining neuroendocrine tuft lineage features marks a paradigm shift, highlighting the plasticity inherent in cancer cells and the potential to intercept malignant evolution by modulating lineage fate.

Subject of Research:
Small-cell lung cancer lineage plasticity driven by genetic alterations and cell of origin.

Article Title:
Basal cell of origin resolves neuroendocrine–tuft lineage plasticity in cancer.

Article References:
Ireland, A.S., Xie, D.A., Hawgood, S.B. et al. Basal cell of origin resolves neuroendocrine–tuft lineage plasticity in cancer. Nature (2025). https://doi.org/10.1038/s41586-025-09503-z

Image Credits:
AI Generated