Extranodal extension (ENE) is a well-established poor prognosticator and an indication for adjuvant treatment escalation in patients with head and neck squamous cell carcinoma (HNSCC). Identification ...of ENE on pretreatment imaging represents a diagnostic challenge that limits its clinical utility. We previously developed a deep learning algorithm that identifies ENE on pretreatment computed tomography (CT) imaging in patients with HNSCC. We sought to validate our algorithm performance for patients from a diverse set of institutions and compare its diagnostic ability to that of expert diagnosticians.
We obtained preoperative, contrast-enhanced CT scans and corresponding pathology results from two external data sets of patients with HNSCC: an external institution and The Cancer Genome Atlas (TCGA) HNSCC imaging data. Lymph nodes were segmented and annotated as ENE-positive or ENE-negative on the basis of pathologic confirmation. Deep learning algorithm performance was evaluated and compared directly to two board-certified neuroradiologists.
A total of 200 lymph nodes were examined in the external validation data sets. For lymph nodes from the external institution, the algorithm achieved an area under the receiver operating characteristic curve (AUC) of 0.84 (83.1% accuracy), outperforming radiologists' AUCs of 0.70 and 0.71 (
= .02 and
= .01). Similarly, for lymph nodes from the TCGA, the algorithm achieved an AUC of 0.90 (88.6% accuracy), outperforming radiologist AUCs of 0.60 and 0.82 (
< .0001 and
= .16). Radiologist diagnostic accuracy improved when receiving deep learning assistance.
Deep learning successfully identified ENE on pretreatment imaging across multiple institutions, exceeding the diagnostic ability of radiologists with specialized head and neck experience. Our findings suggest that deep learning has utility in the identification of ENE in patients with HNSCC and has the potential to be integrated into clinical decision making.
Identification of nodal metastasis and tumor extranodal extension (ENE) is crucial for head and neck cancer management, but currently only can be diagnosed via postoperative pathology. Pretreatment, ...radiographic identification of ENE, in particular, has proven extremely difficult for clinicians, but would be greatly influential in guiding patient management. Here, we show that a deep learning convolutional neural network can be trained to identify nodal metastasis and ENE with excellent performance that surpasses what human clinicians have historically achieved. We trained a 3-dimensional convolutional neural network using a dataset of 2,875 CT-segmented lymph node samples with correlating pathology labels, cross-validated and fine-tuned on 124 samples, and conducted testing on a blinded test set of 131 samples. On the blinded test set, the model predicted ENE and nodal metastasis each with area under the receiver operating characteristic curve (AUC) of 0.91 (95%CI: 0.85-0.97). The model has the potential for use as a clinical decision-making tool to help guide head and neck cancer patient management.
Invadopodia are branched actin-rich structures associated with extracellular matrix (ECM) degradation that collectively form the invasive machinery of aggressive cancer cells. Cortactin is a ...prominent component and a specific marker of invadopodia. Amplification of cortactin is associated with poor prognosis in head and neck squamous cell carcinomas (HNSCC), possibly because of its activity in invadopodia. Although the role of cortactin in invadopodia has been attributed to signaling and actin assembly, it is incompletely understood. We made HNSCC cells deficient in cortactin by RNA interference knockdown methods. In these cortactin knockdown cells, invadopodia were reduced in number and lost their ability to degrade ECM. In the reverse experiment, overexpression of cortactin dramatically increased ECM degradation, far above and beyond the effect on formation of actin/Arp3-positive invadopodia puncta. Secretion of matrix metalloproteinases (MMP) MMP-2 and MMP-9, as well as plasma membrane delivery of MT1-MMP correlated closely with cortactin expression levels. MMP inhibitor treatment of control cells mimicked the cortactin knockdown phenotype, with abolished ECM degradation and fewer invadopodia, suggesting a positive feedback loop in which degradation products from MMP activity promote new invadopodia formation. Collectively, these data suggest that a major role of cortactin in invadopodia is to regulate the secretion of MMPs and point to a novel mechanism coupling dynamic actin assembly to the secretory machinery, producing enhanced ECM degradation and invasiveness. Furthermore, these data provide a possible explanation for the observed association between cortactin overexpression and enhanced invasiveness and poor prognosis in HNSCC patients.
The prognostication of head and neck squamous cell carcinoma (HNSCC) is largely based upon the tumor size and location and the presence of lymph node metastases. Here we show that gene expression ...patterns from 60 HNSCC samples assayed on cDNA microarrays allowed categorization of these tumors into four distinct subtypes. These subtypes showed statistically significant differences in recurrence-free survival and included a subtype with a possible EGFR-pathway signature, a mesenchymal-enriched subtype, a normal epithelium-like subtype, and a subtype with high levels of antioxidant enzymes. Supervised analyses to predict lymph node metastasis status were approximately 80% accurate when tumor subsite and pathological node status were considered simultaneously. This work represents an important step toward the identification of clinically significant biomarkers for HNSCC.
Objectives/Hypothesis
The objective was to characterize incidence, treatment, and survival for hypopharyngeal cancer in the United States between 1988 and 2010, and to analyze associations between ...changes in treatment modality and survival.
Study Design
Retrospective cohort study.
Methods
A total of 3,958 adult patients with hypopharyngeal cancer were identified in the Surveillance, Epidemiology, and End Results database. Incidence, treatment, and survival, controlling for patient demographics and disease severity, were analyzed using two‐tailed t tests, Kaplan‐Meier analysis, and univariate and multivariate Cox regression.
Results
The incidence of hypopharyngeal cancer decreased from 1973 to 2010 with an average annual percent change (APC) of −2.0% every year (P < .05). Treatment with laryngopharyngectomy decreased (−2.5% APC, P < .001), treatment with radiotherapy without surgery increased (+2.0% APC, P < .001), and treatment with neither surgery nor radiotherapy increased (+0.5% APC, P < .001) between 1988 and 2010. There was a significant increase in the 5‐year overall survival between 1988 and 1990 and between 1991 and 1995 (P = .024) with no other significant temporal trends in survival. Multivariate analysis revealed that age (65–74, 75–84, or 85+ relative to 18–54 years old), race (white relative to non‐African races), T stage (T2, T3, or T4 relative to T1), N stage (N2 or N3 relative to N0), and treatment modality (−surgery/−radiation, −surgery/+radiation, and +surgery/−radiation relative to +surgery/+radiation) were all significantly associated with worse survival.
Conclusions
Hypopharyngeal cancer has had a decreasing incidence with little change in patient or tumor characteristics. Treatment has increasingly involved radiation without laryngopharyngectomy. This has not been associated with a decrease in survival. Controlling for patient demographics and disease severity, radiation with laryngopharyngectomy is associated with improved survival.
Level of Evidence
2b Laryngoscope, 124:2064–2069, 2014
We hypothesized that chronic inhibition of epidermal growth factor receptor (EGFR) by cetuximab, a monoclonal anti-EGFR antibody, induces up-regulation of its ligands resulting in resistance and that ...microRNAs (miRs) play an important role in the ligand regulation in head and neck squamous cell carcinoma (HNSCC).
Genome-wide changes in gene and miR expression were determined in cetuximab-sensitive cell line, SCC1, and its resistant derivative 1Cc8 using DNA microarrays and RT-PCR. The effects of differentially expressed EGFR ligands and miRs were examined by MTS, colony formation, ELISA, and western blot assays. Heparin-binding EGF-like growth factor (HB-EGF) and its regulator, miR-212, were differentially expressed with statistical significance when SCC1 and 1Cc8 were compared for gene and miR expression. Stimulation with HB-EGF induced cetuximab resistance in sensitive cell lines. Inhibition of HB-EGF and the addition of miR-212 mimic induced cetuximab sensitivity in resistant cell lines. MicroRNA-212 and HB-EGF expression were inversely correlated in an additional 33 HNSCC and keratinocyte cell lines. Six tumors and 46 plasma samples from HNSCC patients were examined for HB-EGF levels. HB-EGF plasma levels were lower in newly diagnosed HNSCC patients when compared to patients with recurrent disease.
Increased expression of HB-EGF due to down-regulation of miR-212 is a possible mechanism of cetuximab resistance. The combination of EGFR ligand inhibitors or miR modulators with cetuximab may improve the clinical outcome of cetuximab therapy in HNSCC.
The
INK4a-ARF locus encodes two unrelated proteins that both function in tumor suppression. p16
INK4a
binds to and inhibits the activity of CDK4 and CDK6, and ARF arrests the cell cycle in a ...p53-dependent manner. We show here that ARF binds to MDM2 and promotes the rapid degradation of MDM2. This interaction is mediated by the exon 1β–encoded N-terminal domain of ARF and a C-terminal region of MDM2. ARF-promoted MDM2 degradation is associated with MDM2 modification and concurrent p53 stabilization and accumulation. The functional consequence of ARF-regulated p53 levels via MDM2 proteolysis is evidenced by the ability of ectopically expressed ARF to restore a p53-imposed G1 cell cycle arrest that is otherwise abrogated by MDM2. Thus, deletion of the
ARF-INK4a locus simultaneously impairs both the INK4a–cyclin D/CDK4-RB and the ARF-MDM2-p53 pathways.