Radiomic signatures reveal multiscale intratumor heterogeneity associated with tissue tolerance and survival in re-irradiated nasopharyngeal carcinoma: a multicenter study.

BACKGROUND: Post-radiation nasopharyngeal necrosis (PRNN) is a severe adverse event following re-radiotherapy for patients with locally recurrent nasopharyngeal carcinoma (LRNPC) and associated with decreased survival. Biological heterogeneity in recurrent tumors contributes to the different risks of PRNN. Radiomics can be used to mine high-throughput non-invasive image features to predict clinical outcomes and capture underlying biological functions. We aimed to develop a radiogenomic signature for the pre-treatment prediction of PRNN to guide re-radiotherapy in patients with LRNPC. METHODS: This multicenter study included 761 re-irradiated patients with LRNPC at four centers in NPC endemic area and divided them into training, internal validation, and external validation cohorts. We built a machine learning (random forest) radiomic signature based on the pre-treatment multiparametric magnetic resonance images for predicting PRNN following re-radiotherapy. We comprehensively assessed the performance of the radiomic signature. Transcriptomic sequencing and gene set enrichment analyses were conducted to identify the associated biological processes. RESULTS: The radiomic signature showed discrimination of 1-year PRNN in the training, internal validation, and external validation cohorts (area under the curve (AUC) 0.713-0.756). Stratified by a cutoff score of 0.735, patients with high-risk signature had higher incidences of PRNN than patients with low-risk signature (1-year PRNN rates 42.2-62.5% vs. 16.3-18.8%, P < 0.001). The signature significantly outperformed the clinical model (P < 0.05) and was generalizable across different centers, imaging parameters, and patient subgroups. The radiomic signature had prognostic value concerning its correlation with PRNN-related deaths (hazard ratio (HR) 3.07-6.75, P < 0.001) and all causes of deaths (HR 1.53-2.30, P < 0.01). Radiogenomics analyses revealed associations between the radiomic signature and signaling pathways involved in tissue fibrosis and vascularity. CONCLUSIONS: We present a radiomic signature for the individualized risk assessment of PRNN following re-radiotherapy, which may serve as a noninvasive radio-biomarker of radiation injury-associated processes and a useful clinical tool to personalize treatment recommendations for patients with LANPC.

A KdV-SIR equation and its analytical solutions: An application for COVID-19 data analysis.

To describe the time evolution of infected persons associated with an epidemic wave, we recently derived the KdV-SIR equation that is mathematically identical to the Kortewegde Vries (KdV) equation in the traveling wave coordinate and that represents the classical SIR model under a weakly nonlinear assumption. This study further discusses the feasibility of applying the KdV-SIR equation and its analytical solutions together with COVID-19 data in order to estimate a peak time for a maximum number of infected persons. To propose a prediction method and to verify its performance, three types of data were generated based on COVID-19 raw data, using the following procedures: (1) a curve fitting package, (2) the empirical mode decomposition (EMD) method, and (3) the 28-day running mean method. Using the produced data and our derived formulas for ensemble forecasts, we determined various estimates for growth rates, providing outcomes for possible peak times. Compared to other methods, our method mainly relies on one parameter, σo (i.e., a time independent growth rate), which represents the collective impact of a transmission rate (ß) and a recovery rate (ν). Utilizing an energy equation that describes the relationship between the time dependent and independent growth rates, our method offers a straightforward alternative for estimating peak times in ensemble predictions.

Impact of salvage radiotherapy on survival of patients with advanced locally recurrent nasopharyngeal carcinoma: Derivation and validation of a predictive model.

BACKGROUND: Salvage radiotherapy (RT) is a potentially curative approach for advanced locally recurrent nasopharyngeal carcinoma (NPC), but it is associated with severe toxicities. We aimed to develop a model to predict which patients would benefit from salvage RT. METHODS: A total of 809 patients who were diagnosed with advanced locally recurrent NPC and treated with salvage RT or palliative chemotherapy (CT) at a high-volume cancer center were included. Patients were randomly split into a training and validation set and matched using inverse probability of treatment weighting. The primary outcome was overall survival (OS). Candidate variables associated with heterogeneous treatment effects were identified with interaction terms in Cox model and incorporated into Salvage Radiotherapy Outcome Score (SARTOS). RESULTS: The final model included five interaction terms indicating that female sex, presence of prior RT-induced grade ≥ 3 late toxicities and suboptimal performance status were associated with less benefit from salvage RT. SARTOS from the model significantly predicted treatment effects of salvage RT in matched training (Pinteration < 0.001) and validation cohorts (Pinteration = 0.027). Of patients in high SARTOS subgroup, salvage RT significantly improved survival versus palliative CT in matched training (3-year OS 67.3% vs. 42.0%, HR 0.51, 95% CI 0.32-0.82, P = 0.005) and validation cohorts (3-year OS 71.8% vs. 22.8%, HR 0.40, 95% CI 0.17-0.97, P = 0.042); in low SARTOS subgroup, salvage RT failed to induce survival benefit. CONCLUSIONS: We found that the SARTOS model could identify a subgroup of patients who benefit from salvage RT versus palliative CT, which helps personalize treatment recommendations for patients with recurrent NPC.

Management of suboptimal response to induction chemotherapy in locoregionally advanced nasopharyngeal carcinoma: Re-induction therapy or direct to Radiotherapy?

BACKGROUND: Unsatisfactory tumor response to induction chemotherapy (IC) is an adverse prognostic factor of locoregionally advanced nasopharyngeal carcinoma (LANPC). A re-induction strategy which applies additional cycles of an alternative IC regimen prior to radiotherapy (RT) has been adopted. METHODS: A total of 419 LANPC patients who attained suboptimal response (stable disease or disease progression) according to the Response Evaluation in Solid Tumors (RECIST) guideline after initial IC were retrospectively included. They were divided into those who received additional cycles of re-induction regimen prior to RT (re-induction group, n = 87) and those who had no additional chemotherapy (direct to RT group, n = 332). Propensity score matching (PSM) was used to adjust for potential confounders. Tumor response and long-term survival were compared between two groups. RESULTS: After receiving a second IC regimen, 39.1% of the patients in re-induction group attained partial response; however, the tumor control of subsequent RT was not significantly improved when compared with direct to RT group (patients attaining complete response after RT 55.2% vs. 52.5%, P = 0.757). Patients who received re-induction therapy showed worse locoregional relapse-free survival (LRFS) and progression-free survival (PFS) than those proceeded directly to RT (3-year LRFS 75.7% vs. 83.1%, P = 0.005; 3-year PFS 62.4% vs. 68.3%, P = 0.037). The increased hematological toxicities were observed in re-induction group that included grade 3-4 anemia, thrombocytopenia and liver enzyme increase. CONCLUSION: Re-induction therapy decreased LRFS and PFS and increased toxicities among patients who attain suboptimal response to initial IC regimen, as compared with direct to RT strategy.