Kelly Y. P. Liu

Wide-field in vivo oral OCT imaging.

We have built a polarization-sensitive swept source Optical Coherence Tomography (OCT) instrument capable of wide-field in vivo imaging in the oral cavity. This instrument uses a hand-held side-looking fiber-optic rotary pullback catheter that can cover two dimensional tissue imaging fields approximately 2.5 mm wide by up to 90 mm length in a single image acquisition. The catheter spins at 100 Hz with pullback speeds up to 15 mm/s allowing imaging of areas up to 225 mm(2) field-of-view in seconds. A catheter sheath and two optional catheter sheath holders have been designed to allow imaging at all locations within the oral cavity. Image quality of 2-dimensional image slices through the data can be greatly enhanced by averaging over the orthogonal dimension to reduce speckle. Initial in vivo imaging results reveal a wide-field view of features such as epithelial thickness and continuity of the basement membrane that may be useful in clinic for chair-side management of oral lesions.

Prospects in the Application of Photodynamic Therapy in Oral Cancer and Premalignant Lesions

Oral cancer is a global health burden with significantly poor survival, especially when the diagnosis is at its late stage. Despite advances in current treatment modalities, there has been minimal improvement in survival rates over the last five decades. The development of local recurrence, regional failure, and the formation of second primary tumors accounts for this poor outcome. For survivors, cosmetic and functional compromises resulting from treatment are often devastating. These statistics underscore the need for novel approaches in the management of this deadly disease. Photodynamic therapy (PDT) is a treatment modality that involves administration of a light-sensitive drug, known as a photosensitizer, followed by light irradiation of an appropriate wavelength that corresponds to an absorbance band of the sensitizer. In the presence of tissue oxygen, cytotoxic free radicals that are produced cause direct tumor cell death, damage to the microvasculature, and induction of inflammatory reactions at the target sites. PDT offers a prospective new approach in controlling this disease at its various stages either as a stand-alone therapy for early lesions or as an adjuvant therapy for advanced cases. In this review, we aim to explore the applications of PDT in oral cancer therapy and to present an overview of the recent advances in PDT that can potentially reposition its utility for oral cancer treatment.

A Robust Protocol for Using Multiplexed Droplet Digital PCR to Quantify Somatic Copy Number Alterations in Clinical Tissue Specimens.

The ability of droplet digital PCR (ddPCR) to accurately determine the concentrations of amplifiable targets makes it a promising platform for measuring copy number alterations (CNAs) in genomic biomarkers. However, its application to clinical samples, particularly formalin-fixed paraffin-embedded specimens, will require strategies to reliably determine CNAs in DNA of limited quantity and quality. When applied to cancerous tissue, those methods must also account for global genetic instability and the associated probability that the abundance(s) of one or more chosen reference loci do not represent the average ploidy of cells comprising the specimen. Here we present an experimental design strategy and associated data analysis tool that enables accurate determination of CNAs in a panel of biomarkers using multiplexed ddPCR. The method includes strategies to optimize primer and probes design to cleanly segregate droplets in the data output from reaction wells amplifying multiple independent templates, and to correct for bias from artifacts such as DNA fragmentation. We demonstrate how a panel of reference loci can be used to determine a stable CNA-neutral benchmark. These innovations, when taken together, provide a comprehensive strategy that can be used to reliably detect biomarker CNAs in DNA extracted from either frozen or FFPE tissue biopsies.

Detection of clinically relevant copy number alterations in oral cancer progression using multiplexed droplet digital PCR

Copy number alterations (CNAs), a common genomic event during carcinogenesis, are known to affect a large fraction of the genome. Common recurrent gains or losses of specific chromosomal regions occur at frequencies that they may be considered distinctive features of tumoral cells. Here we introduce a novel multiplexed droplet digital PCR (ddPCR) assay capable of detecting recurrent CNAs that drive tumorigenesis of oral squamous cell carcinoma. Applied to DNA extracted from oral cell lines and clinical samples of various disease stages, we found good agreement between CNAs detected by our ddPCR assay with those previously reported using comparative genomic hybridization or single nucleotide polymorphism arrays. Furthermore, we demonstrate that the ability to target specific locations of the genome permits detection of clinically relevant oncogenic events such as small, submicroscopic homozygous deletions. Additional capabilities of the multiplexed ddPCR assay include the ability to infer ploidy level, quantify the change in copy number of target loci with high-level gains, and simultaneously assess the status and viral load for high-risk human papillomavirus types 16 and 18. This novel multiplexed ddPCR assay therefore may have clinical value in differentiating between benign oral lesions from those that are at risk of progressing to oral cancer.

Altered Immune-Related Gene Expressions Indicate Oral Cancer Nodal Disease

Lymph nodal disease (LN+) is the most significant prognostic factor of oral squamous cell carcinoma (OSCC). Current risk indicator(s) for guiding elective neck dissection (END) is insufficient for clinically node-negative (cN0) patients, resulting in under- or overtreatment. While the role of immunological events in tumorigenesis and metastasis is evident, the prognostic implication in OSCC remains unclear. The study objective was to investigate large-scale immune-related gene expression and determine its prognostic value on node-free survival (NFS). We analyzed patients who received intent-to-cure surgery with at least 3 y of follow-up and known outcome of LN through a pan-Canadian surgical trial. Total RNA was extracted from surgical tissues with >70% tumor content and analyzed on a 730-gene panel (NanoString nCounter® PanCancer Immune Panel). We first profiled gene expression in a fresh-frozen (FF) discovery set to identify differentially expressed (DE) genes, which were then used in unsupervised clustering analysis to identify patient subgroups. The prognostic value of the identified DE genes was then validated on formalin-fixed, paraffin-embedded (FFPE) samples. A total of 177 RNA samples were derived from 89 FF and 88 FFPE surgical tissues, of which 45 (51%) and 40 (45%), respectively, were from patients who developed LN+. We identified 6 DE genes overexpressed in LN+ tumors (false discovery rate <0.001; log2 fold change >1). Clustering analysis separated the patients into 2 subgroups (CM1, CM2), with CM2 exhibiting significantly increased expression and worse 5-y NFS rate (28%; P < 0.001). The prognostic value of these 6 candidate genes was validated on FFPE samples, which were also separated into 2 distinct prognostic groups, confirming the association between increased gene expression and poor 5-y NFS (CM1, 70.3%; CM2, 43.3%; P = 0.01). This is the first study identifying a panel of immune-related genes associated with NFS that can potentially be used clinically stratifying the risk of LN+ at the time of OSCC diagnosis.

Wide-field OCT imaging of oral lesions in vivo: quantification and classification (Conference Presentation)

Worldwide, there are over 450,000 new cases of oral cancer reported each year. Late-stage diagnosis remains a significant factor responsible for its high mortality rate (>50%). In-vivo non-invasive rapid imaging techniques, that can visualise clinically significant changes in the oral mucosa, may improve the management of oral cancer. We present an analysis of features extracted from oral images obtained using our hand- held wide-field Optical Coherence Tomography (OCT) instrument. The images were analyzed for epithelial scattering, overall tissue scattering, and 3D basement membrane topology. The associations between these three features and disease state (benign, pre-cancer, or cancer), as measured by clinical assessment or pathology, were determined. While scattering coefficient has previously been shown to be sensitive to cancer and dysplasia, likely due to changes in nuclear and cellular density, the addition of basement membrane topology may increase diagnostic ability- as it is known that the presence of bulbous rete pegs in the basement membrane are characteristic of dysplasia. The resolution and field-of-view of our oral OCT system allowed analysis of these features over large areas of up to 2.5mm x 90mm, in a timely fashion, which allow for application in clinical settings.

An actionable test using loss of heterozygosity in identifying high-risk oral premalignant lesions

OBJECTIVES To develop an actionable test using fluorescence capillary electrophoresis (FCE) to assess loss of heterozygosity (LOH) of histologically similar low-grade lesions (LGLs) to identify high-risk lesions for oral cancer progression. STUDY DESIGN To determine the cutoffs of LOH, the FCE results of 52 surgical margin samples were used to compare with the existing LOH results from the previously validated 32 P-GE approach. Using the developed FCE workflow, an independent set of 102 LGLs with known progression status was used to determine the LOH molecular risk (MR) patterns and associated risk of progression. RESULTS Using 65% cutoff LOH-FCE, the agreement of LOH-32 P-GE had an average of 82.3% (76.8-87.8). Compared with nonprogressors (n = 61), anatomic site and MR patterns (LOH at 9 p21, 3 p14, or 17 p13) were independent risk factors. High-risk profile of tongue and MR3 (LOH at 9 p21 and/or 3 p14 and 17 p13) was significantly associated with progression (hazard ratio [HR] 6.7; 95% confidence interval [CI] 2.6-17.6) with specificity of 98.4% at identifying progressors. CONCLUSIONS We have developed an objective test using LOH to stratify the risk of LGLs. With further validation, it can be used in the clinical settings to provide clinicians additional information guiding the management of these lesions.

Novel computational image analysis to predict regional nodal disease for early-stage oral cancer

Objective: Nodal disease (N+) for early-stage oral squamous cell carcinoma (OSCC) is the most significant prognostic factor for survival. There is a lack of effective predictor to justify prophylactic neck dissection. Quantitative tissue pathology (QTP) has shown its promise in providing an objective means for diagnosis and prognosis of many cancer types. We conducted a pilot study on the utilization of QTP to evaluate risk of nodal disease. Study design: Retrospective case-control study Subjects and methods: Histological sections from 15 primary tumors of clinically node-negative (N0) patients were stained with Feulgen-Thionin followed by acquisition of digital images and image processing to measure the mean and variance of nuclear phenotype and tissue architecture features from 45,253 nuclei of 45 tumor nests. Association between features and nodal disease outcome (N0 or N+) was investigated using nested analysis of variance adjusted by patient. Ability to discriminate between N0 and N+ was analyzed using multivariate logistic regression and receiver operating characteristics (ROC) curve analysis. P-value<0.05 (2-sided) was considered significant. Results: The N+ group presented higher mean values of chromatin condensation levels and cell density compared to those of the N0 group. ROC curve showed a strong discriminative ability of chromatin condensation levels between the N+ and N0 groups with a sensitivity and specificity of 1.0 and 0.75, respectively. Conclusion: This study reports the first-ever data on QTP as a risk assessment tool for nodal disease in early-stage OSCCs. Such computational imaging analysis potentially provides a new objective approach to predict regional nodal disease. Correspondence to: Dr. Catherine F. Poh, Department of Oral Medical and Biological Sciences, Faculty of Dentistry, The University of British Columbia, 2199 Wesbrook Mall, Vancouver BC, V6T 1Z3, Canada; E-mail: cpoh@dentistry.ubc.ca

Adverse Events Associated with Postsurgery of Early-Stage Oral Cancer in a Phase III Surgical Trial

Objectives: 1) Investigate the adverse events (AEs) associated with surgery for early-stage oral cancer. 2) Characterize surgical procedure and its related AEs. Methods: Based on the National Institute of Health/National Cancer Institute Common Terminology Criteria for Adverse Events (v4.0), patient self-reported AEs post-surgery in a pan-Canadian multi-center phase III randomized controlled surgical trial (the COOLS trial) were captured during the follow-up visits. Results: From September 2010 to January 2013, 228 eligible patients were recruited and received surgeries, including tumor excision, and/or free flap or neck dissection. Among these, 188 (82%) patients had at least 3 months follow-up with an average of 24.8±19 months. There were 83 types and 593 counts of AEs from 148 (79%) patients. The most occurring AEs were oral pain (146 patients) and oral dysesthesia (64 patients). Thirty patients (20%) presented with prolonged oral pain (≥3 months) with an average duration of 6.5±10.5 months for Grade I and 2.0±5.6 months for Grade 2 or greater (77 patients). The average duration of oral dysethesia was 18.6±16.4 months with 52 (80%) over 3 months. Univariate analysis had shown that neck dissection and longer surgery time have significant impact on the reported Grade 2 oral pain (P = 0.03 for both). Interestingly, excision type [cold knife (9%) vs. electric knife/laser (23%)] had significant effect on experiencing prolonged post-surgical oral pain (P = 0.01). Conclusions: Post-surgical AEs surveillance is critical to understand the factors impacting patients’ quality of life. Concurrent neck dissection has impacts on patients’ experience in prolonged post-surgical oral pain.

Abstract 5125: Predictive markers on risk of developing lymph node metastasis in early-staged oral cancer patients.

Proceedings: AACR 104th Annual Meeting 2013; Apr 6-10, 2013; Washington, DC Introduction: Cervical lymph nodal metastasis is the most important variable for unimproved survival of oral squamous cell carcinoma (OSCC) patients. Resulting from lack of predictive marker, unnecessary neck dissection (ND) has tremendous impact on costs and patients quality of life. Current literature have shown contradictory results in predicting markers using clinico-pathological parameters. Objectives: 1) to collect the demographics, clinico-pathological information of primary OSCC patients with intent-to-cure surgery; 2) to categorize these patients according to their nodal status at and after surgery; 3) to determine the impact of nodal status to overall survival; and 4) to assess clinic-pathological variables predicting nodal disease of N0 early-stage OSCC patients. Methods: Between 2003 and 2007, 303 primary OSCC patients were identified from the BC Cancer Registry Database with complete clinico-pathological information and received primary curative surgical treatment with at least 5-year follow-up (FU). Patients were categorized into 4 groups: Gr.A were N0 at surgery or during FU (N=118); Gr.B were N0 at surgery but N+ during FU (N=57); Gr.C received concurrent ND and were N0 at the time of surgery (N=57); and Gr.D received concurrent ND and were N+ at the time of surgery (N=71). Data retrieved included demographics, clinico-pathological factors, treatment, and time to outcomes (survival or nodal disease). Results: Nodal disease at the time of surgery (Gr.D) or during the FU (Gr.B) has a pivotal impact on the 5-year survival rates (44% and 55%, respectively, P<0.0001). Cox proportional hazard models identified positive nodal disease, TNM staging, and adjuvant radiotherapy as significant predictors for survival (HR: 2.43, 95%CI, 1.18-5.02; P=0.02; 1.31, 95%CI, 1.0-1.7; P=0.03; 6.53, 95%CI, 2.86-14.91; P<0.0001, respectively). Among the 205 N0 at the time of surgery (Gr.A+Gr.B), strikingly, one-in-four developed nodal disease in average 13.2±14.1 months and with 76% in the first 18-months post surgery. Between Gr.A and Gr.B, there was no significant differences in tumor depth (4.2±0.4 mm vs. 5.0±0.5 mm, P=0.25). Using multivariate analysis, age (P=0.03) and tumor grade (P=0.001) were significant predictive markers. Tumor depth of 4 mm, the current standard factor on the necessity of prophylactic ND, did not predict nodal status (P=0.11). Conclusion: Nodal status is highly associated with patient survival. The data strongly suggest aggressiveness of neck metastasis either at the time of surgery or during FU. Effective markers to predict nodal disease pre-surgery can benefit high-risk patients to have early intervention and avoid unnecessary ND for the low-risk. (Supported by the Canadian Cancer Society Research Institute (CCSRI-20336), and Terry Fox Research Institute (TFRI-2009-24). CFP is supported by a Scholar Award from the Michael Smith Foundation for Health Research.) Citation Format: Kelly YP Liu, Scott Durham, Kenneth W. Berean, Catherine F. Poh. Predictive markers on risk of developing lymph node metastasis in early-staged oral cancer patients. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5125. doi:10.1158/1538-7445.AM2013-5125 Note: This abstract was not presented at the AACR Annual Meeting 2013 because the presenter was unable to attend.