Integrating Artificial Intelligence and Machine Learning Into Cancer Clinical Trials

The practice of oncology requires analyzing and synthesizing abundant data. From the patient's workup to determine eligibility to the therapies received to the post-treatment surveillance, practitioners must constantly juggle, evaluate, and weigh decision-making based on their best understanding of information at hand. These complex, multifactorial decisions have a tremendous opportunity to benefit from data-driven machine learning (ML) methods to drive opportunities in artificial intelligence (AI). Within the past 5 years, we have seen AI move from simply a promising opportunity to being used in prospective trials. Here, we review recent efforts of AI in clinical trials that have moved the needle towards improved prediction of actionable outcomes, such as predicting acute care visits, short term mortality, and pathologic extranodal extension. We then pause and reflect on how these AI models ask a different question than traditional statistics models that readers may be more familiar with; how then should readers conceptualize and interpret AI models that they are not as familiar with. We end with what we believe are promising future opportunities for AI in oncology, with an eye towards allowing the data to inform us through unsupervised learning and generative models, rather than asking AI to perform specific functions.

Deep Learning for Fully Automated Prediction of Overall Survival in Patients Undergoing Resection for Pancreatic Cancer: A Retrospective Multicenter Study

OBJECTIVE

To develop an imaging-derived biomarker for prediction of overall survival (OS) of pancreatic cancer by analyzing preoperative multiphase contrast-enhanced computed topography (CECT) using deep learning.

BACKGROUND

Exploiting prognostic biomarkers for guiding neoadjuvant and adjuvant treatment decisions may potentially improve outcomes in patients with resectable pancreatic cancer.

METHODS

This multicenter, retrospective study included 1516 patients with resected pancreatic ductal adenocarcinoma (PDAC) from 5 centers located in China. The discovery cohort (n=763), which included preoperative multiphase CECT scans and OS data from 2 centers, was used to construct a fully automated imaging-derived prognostic biomarker-DeepCT-PDAC-by training scalable deep segmentation and prognostic models (via self-learning) to comprehensively model the tumor-anatomy spatial relations and their appearance dynamics in multiphase CECT for OS prediction. The marker was independently tested using internal (n=574) and external validation cohorts (n=179, 3 centers) to evaluate its performance, robustness, and clinical usefulness.

RESULTS

Preoperatively, DeepCT-PDAC was the strongest predictor of OS in both internal and external validation cohorts [hazard ratio (HR) for high versus low risk 2.03, 95% confidence interval (CI): 1.50-2.75; HR: 2.47, CI: 1.35-4.53] in a multivariable analysis. Postoperatively, DeepCT-PDAC remained significant in both cohorts (HR: 2.49, CI: 1.89-3.28; HR: 2.15, CI: 1.14-4.05) after adjustment for potential confounders. For margin-negative patients, adjuvant chemoradiotherapy was associated with improved OS in the subgroup with DeepCT-PDAC low risk (HR: 0.35, CI: 0.19-0.64), but did not affect OS in the subgroup with high risk.

CONCLUSIONS

Deep learning-based CT imaging-derived biomarker enabled the objective and unbiased OS prediction for patients with resectable PDAC. This marker is applicable across hospitals, imaging protocols, and treatments, and has the potential to tailor neoadjuvant and adjuvant treatments at the individual level.

Novel MSH6 mutation predicted metastasis in eyelid and periocular squamous cell carcinoma

BACKGROUND

Our previous research revealed the relative local aggressiveness of eyelid and periocular squamous cell carcinoma (EPSCC), but its distinct genetic characteristics involved remain unknown.

OBJECTIVES

We conducted this study based on next-generation sequencing to identify the genetic distinctiveness of EPSCC and damaging mutations for possible etiology and poor prognosis.

METHODS

We performed sequencing using a 556-gene panel (smartonco) in 48 EPSCCs. Cox hazards model was applied to explore mutated genes that increase risk of metastasis and death. Pathogenesis of the mutations was predicted by sequence alignment algorithms.

RESULTS

The most commonly mutated genes were KMT2C (N=17, 35%), LRP1B (N=14, 29%), KMT2D (N=12, 25%), PTCH1(N=10, 21%) and TP53(N=10, 21%). DNA mismatch repair (MMR) genes (42%) like MSH6(19%) and MLH3(12%) were among the most frequently mutated genes. Cell cycle regulators including TP53(21%) and CDKN2A (10%) were less frequently mutated than in other squamous cell carcinomas (SCCs). Ultra violet exposure, MMR deficiency and aging were the main etiology. Of note, KMT2C has a deleterious mutation hotspot. Patients burdened with MSH6 mutation has a higher risk of overall metastasis (P=0.045, HR=5.165) and nodal metastasis (P=0.022, HR=14.038). Moreover, a hotspot mutation MSH6^E52A brought an even higher risk of nodal metastasis (P=0.011, HR=18.745).

CONCLUSIONS

EPSCCs displayed a unique mutation profile from cutaneous SCCs and mucosal SCCs. We have identified novel damaging mutations in epigenetic regulators like KMT2C boosted early onset of EPSCCs in addition to UVR, aging or MMR deficiency. And malfunction of MMR genes worsened prognosis.

Artificial Intelligence Surgery: How Do We Get to Autonomous Actions in Surgery?

Most surgeons are skeptical as to the feasibility of autonomous actions in surgery. Interestingly, many examples of autonomous actions already exist and have been around for years. Since the beginning of this millennium, the field of artificial intelligence (AI) has grown exponentially with the development of machine learning (ML), deep learning (DL), computer vision (CV) and natural language processing (NLP). All of these facets of AI will be fundamental to the development of more autonomous actions in surgery, unfortunately, only a limited number of surgeons have or seek expertise in this rapidly evolving field. As opposed to AI in medicine, AI surgery (AIS) involves autonomous movements. Fortuitously, as the field of robotics in surgery has improved, more surgeons are becoming interested in technology and the potential of autonomous actions in procedures such as interventional radiology, endoscopy and surgery. The lack of haptics, or the sensation of touch, has hindered the wider adoption of robotics by many surgeons; however, now that the true potential of robotics can be comprehended, the embracing of AI by the surgical community is more important than ever before. Although current complete surgical systems are mainly only examples of tele-manipulation, for surgeons to get to more autonomously functioning robots, haptics is perhaps not the most important aspect. If the goal is for robots to ultimately become more and more independent, perhaps research should not focus on the concept of haptics as it is perceived by humans, and the focus should be on haptics as it is perceived by robots/computers. This article will discuss aspects of ML, DL, CV and NLP as they pertain to the modern practice of surgery, with a focus on current AI issues and advances that will enable us to get to more autonomous actions in surgery. Ultimately, there may be a paradigm shift that needs to occur in the surgical community as more surgeons with expertise in AI may be needed to fully unlock the potential of AIS in a safe, efficacious and timely manner.

Nodal staging convergence for HPV- and HPV+ oropharyngeal carcinoma

BACKGROUND

Modern disease staging systems have restructured human papillomavirus (HPV)-negative (HPV-) and HPV-positive (HPV+) oropharyngeal carcinoma (OPC) into distinct pathologic nodal systems. Given that quantitative lymph node (LN) burden is the dominant prognostic factor in most head and neck cancers, we investigated whether HPV- and HPV+ OPC warrant divergent pathologic nodal classification.

METHODS

Multivariable Cox regression models of OPC surgical patients identified via U.S. cancer registry data were constructed to determine associations between survival and nodal characteristics. Nonlinear associations between metastatic LN number and survival were modeled with restricted cubic splines. Recursive partitioning analysis (RPA) was used to derive unbiased nodal schema.

RESULTS

Mortality risk escalated continuously with each successive positive LN in both OPC subtypes, with analogous slope. Survival hazard increased by 18.5% (hazard ratio [HR], 1.19 [95% CI, 1.16-1.21]; P < .001) and 19.1% (HR, 1.19 [95% CI, 1.17-1.21]; P < .001), with each added positive LN for HPV- and HPV+ OPC, respectively, up to identical change points of 5 positive LNs. Extranodal extension (ENE) was an independent predictor of HPV- OPC (HR, 1.55 [95% CI, 1.20-1.99]; P < .001) and HPV+ OPC (HR 1.73 [95% CI, 1.36-2.20]; P < .001) mortality. In RPA for both diseases, metastatic LN was the principal nodal covariate driving survival, with ENE as a secondary determinant. Given the similarities across analyses, we propose a concise, unifying HPV-/HPV+ OPC pathologic nodal classification schema: N1, 1-5 LN+/ENE-; N2, 1-5 LN+/ENE+; N3, >5 LN+.

CONCLUSION

HPV- and HPV+ OPC exhibit parallel relationships between nodal characteristics and relative mortality. In both diseases, metastatic LN number represents the principal nodal covariate governing survival, with ENE being an influential secondary element. A consolidated OPC pathologic nodal staging system that is based on these covariates may best convey prognosis.

LAY SUMMARY

The current nodal staging system for oropharyngeal carcinoma (OPC) has divided human papillomavirus (HPV)-negative (HPV-) and HPV-positive (HPV+) OPC into distinct systems that rely upon criteria that establish them as separate entities, a complexity that may undermine the core objective of staging schema to clearly communicate prognosis. Our large-scale analysis revealed that HPV- and HPV+ pathologic nodal staging systems in fact mirror each other. Multiple analyses produced conspicuously similar nodal staging systems, with metastatic lymph node number and extranodal extension delineating the highest risk groups that shape prognosis. We propose unifying HPV- and HPV+ nodal systems to best streamline prognostication and maximize staging accuracy.