[The emulation of clinical trials with real-world data: development and application of target trial].

Clinical trial is the gold standard for evaluating the efficacy and safety of interventions; however, it is limited by high costs and long time. Real-world data (RWD) can provide a robust data basis for comparative research, but the quality is uneven. This review introduces the target trial emulation, in which researchers, using RWD and following the design of clinical trials, define exposure and outcome in advance, set eligibility criteria, determine the time zero, estimate sample size, and plan statistical analysis, to enhance the quality of evidence for observational studies. This review preliminarily discusses the standard of evidence quality evaluation in target trial emulation. Then, the target trial emulation is shown through case interpretation.

Diffusion MRI correlation with p16 status and prediction for tumor progression in locally advanced head and neck cancer.

Purpose: To investigate p16 effects on diffusion image metrics and associations with tumor progression in patients with locally advanced head and neck cancers. Methods: Diffusion images pretreatment and after 20 Gy (2wk) of RT were analyzed in patients with cT4/N3 p16+ oropharynx cancer (OPSCC) (N=51) and locoregionally advanced head and neck squamous cell carcinoma (LAHNSCC) (N=28), enrolled onto a prospective adaptive RT trial. Mean ADC values, subvolumes with ADC <1.2 um2/ms (TVLADC), and peak values of low (µL) and high (µH) components of ADC histograms in primary and total nodal gross tumor volumes were analyzed for prediction of freedom from local, distant, or any progression (FFLP, FFDP or FFLRDP) using multivariate Cox proportional-hazards model with clinical factors. P value with false discovery control <0.05 was considered as significant. Results: With a mean follow up of 36 months, 18 of LAHNSCC patients and 16 of p16+ OPSCC patients had progression. After adjusting for p16, small µL and ADC values, and large TVLADC of primary tumors pre-RT were significantly associated with superior FFLRDP, FFLP and FFDP in the LAHNSCC (p<0.05), but no diffusion metrics were significant in p16+ oropharynx cancers. Post ad hoc analysis of the p16+ OPSCC only showed that large TVLADC of the total nodal burden pre-RT was significantly associated with inferior FFDP (p=0.05). Conclusion: ADC metrics were associated with different progression patterns in the LAHNSCC and p16+ OPSCC, possibly explained by differences in cancer biology and morphology. A deep understanding of ADC metrics is warranted to establish imaging biomarkers for adaptive RT in HNSCC.