Understanding of dedifferentiation, an indicator of poo prognosis for patients with thyroid cancer, has been hampered by imprecise and incomplete characterization of its heterogeneity and its attributes. Using single-cell RNA sequencing, we explored the landscape of thyroid cancer at single-cell resolution with 46,205 cells and delineated its dedifferentiation process and suppressive immune microenvironment. The developmental trajectory indicated that anaplastic thyroid cancer (ATC) cells were derived from a small subset of papillary thyroid cancer (PTC) cells. Moreover, a potential functional role of CREB3L1 on ATC development was revealed by integrated analyses of copy number alteration and transcriptional regulatory network. Multiple genes in differentiation-related pathways (e.g., EMT) were involved as the downstream targets of CREB3L1, increased expression of which can thus predict higher relapse risk of PTC. Collectively, our study provided insights into the heterogeneity and molecular evolution of thyroid cancer and highlighted the potential driver role of CREB3L1 in its dedifferentiation process.

PURPOSE: Primary or acquired resistance to cetuximab, an antiepidermal growth factor receptor monoclonal antibody (mAb), minimizes its utility in recurrent/metastatic head and neck squamous cell carcinoma (HNSCC). Aberrant hepatocyte growth factor/cMet pathway activation is an established resistance mechanism. Dual pathway targeting may overcome resistance. PATIENTS AND METHODS: This multicenter, randomized, noncomparative phase II study evaluated ficlatuzumab, an antihepatocyte growth factor mAb, with or without cetuximab in recurrent/metastatic HNSCC. The primary end point was median progression-free survival (PFS); an arm met significance criteria if the lower bound of the 90% CI excluded the historical control of 2 months. Key eligibility criteria were HNSCC with known human papillomavirus (HPV) status, cetuximab resistance (progression within 6 months of exposure in the definitive or recurrent/metastatic setting), and resistance to platinum and anti–PD-1 mAb. Secondary end points included objective response rate (ORR), toxicity, and the association of HPV status and cMet overexpression with efficacy. Continuous Bayesian futility monitoring was used. RESULTS: From 2018 to 2020, 60 patients were randomly assigned and 58 were treated. Twenty-seven versus 33 patients were allocated to monotherapy versus combination. Arms were balanced for major prognostic factors. The monotherapy arm closed early for futility. The combination arm met prespecified significance criteria with a median PFS of 3.7 months (lower bound 90% CI, 2.3 months; P = .04); the ORR was 6 of 32 (19%), including two complete and four partial responses. Exploratory analyses were limited to the combination arm: the median PFS was 2.3 versus 4.1 months ( P = .03) and the ORR was 0 of 16 (0%) versus 6 of 16 (38%; P = .02) in the HPV-positive versus HPV-negative subgroups, respectively. cMet overexpression was associated with reduced hazard of progression in HPV-negative but not HPV-positive disease ( P interaction = .02). CONCLUSION: The ficlatuzumab-cetuximab arm met significance criteria for PFS and warrants phase III development. HPV-negative HNSCC merits consideration as a selection criterion.

Oral squamous cell carcinoma (OSCC) develops through the multistep malignant progression of squamous epithelium. This process can be prevented by PD-1 blockade in a mouse model for oral carcinogenesis. OSCCs exhibit a high incidence of p53 mutations that confer oncogenic gain-of-function (GOF) activities that promote resistance to standard therapies and poor clinical outcomes. To determine whether epithelial p53 mutations modulate anti-PD-1-mediated oral cancer immunoprevention, we generated mouse models for oral carcinogenesis by exposing mice carrying epithelial-specific p53 mutations to the carcinogen 4NQO. Consistent with the oncogenic functions of mutant p53, mice with OSCCs expressing the p53^R172H GOF mutation developed higher metastasis rates than mice with loss-of-function (LOF) p53 deletion or with wild-type p53. Throughout oral cancer progression, pre-invasive and invasive lesions showed a gradual increase in T-cell infiltration, recruitment of immunosuppressive regulatory T-cells (Tregs), and induction of PD-1/PD-L1 immune checkpoint proteins. Notably, while PD-1 blockade prevented the development of OSCCs in mice with wild-type p53 or p53 deletion, GOF p53^R172H abrogated the immunopreventive effects of anti-PD-1, associated with upregulation of IL17 signaling and depletion of exhausted CD8 cells in the microenvironment of the p53^R172H tumors. These findings sustain a potential role for p53 profiling in personalized oral cancer immunoprevention.

Head and neck squamous cell carcinoma (HNSCC) constitutes 90% of head and neck cancers. HNSCC development is linked to chronic inflammation, while established HNSCC tumors are often immune suppressive. However, both occur through mechanisms that are not fully understood. The cytosolic double-stranded DNA sensor Absent in Melanoma 2 (AIM2) is an inflammasome forming protein that also has inflammasome-distinct roles in restricting tumorigenesis by limited PI3K signaling. Here, we used an experimental mouse model of HNSCC, involving treatment of wild type (WT) and Aim2 ^−/− mice with the carcinogen 4NQO in drinking water. Compared to WT mice, 4NQO-treated Aim2 ^−/− mice exhibited larger tumor sizes and increased tissue dysplasia. 4NQO-treated wild type and Aim2 ^−/− mice displayed similar tongue Il6, Tnf, Il1b, Il12, and Il10 expression and no consistent differences in PI3K or inflammasome activation, suggesting AIM2 may not regulate these factors during HNSCC. Instead, Ifng and Irf1 was elevated in 4NQO-treated Aim2 ^−/− mice, suggesting AIM2 restricts IFNγ. In line with this, RNA-sequencing of total tongue RNA from 4NQO-treated mice revealed Aim2 ^−/− mice had enhanced expression of genes related to the MHC protein complex, cell killing, and T cell activation compared to wild type mice. In addition, we observed increased macrophage infiltration into the tongue epithelium of 4NQO-treated Aim2 ^−/− mice. Lastly, using Aim2 ^−/− / Rag1 ^−/− -double deficient animals, we found that the adaptive immune compartment was necessary for the enhanced tumorigenesis during AIM2 deficiency. Taken together, these findings suggest AIM2 limits the progression of oral tumor development partially through regulating IFNγ and adaptive immune responses.