Background and Aims: Current prediction models for early recurrence of hepatocellular carcinoma (HCC) after surgical resection remain unsatisfactory. The aim of this study was to develop evolutionary learning-derived prediction models with interpretability using both clinical and radiomic features to predict early recurrence of HCC after surgical resection. Methods: Consecutive 517 HCC patients receiving surgical resection with available contrast-enhanced computed tomography (CECT) images before resection were retrospectively enrolled. Patients were randomly assigned to a training set (n = 362) and a test set (n = 155) in a ratio of 7:3. Tumor segmentation of all CECT images including noncontrast phase, arterial phase, and portal venous phase was manually performed for radiomic feature extraction. A novel evolutionary learning-derived method called genetic algorithm for predicting recurrence after surgery of liver cancer (GARSL) was proposed to design prediction models for early recurrence of HCC within 2 years after surgery. Results: A total of 143 features, including 26 preoperative clinical features, 5 postoperative pathological features, and 112 radiomic features were used to develop GARSL preoperative and postoperative models. The area under the receiver operating characteristic curves (AUCs) for early recurrence of HCC within 2 years were 0.781 and 0.767, respectively, in the training set, and 0.739 and 0.741, respectively, in the test set. The accuracy of GARSL models derived from the evolutionary learning method was significantly better than models derived from other well-known machine learning methods or the early recurrence after surgery for liver tumor (ERASL) preoperative (AUC = 0.687, p < 0.001 vs. GARSL preoperative) and ERASL postoperative (AUC = 0.688, p < 0.001 vs. GARSL postoperative) models using clinical features only. Conclusion: The GARSL models using both clinical and radiomic features significantly improved the accuracy to predict early recurrence of HCC after surgical resection, which was significantly better than other well-known machine learning-derived models and currently available clinical models.

The ability to detect cancer at an early stage in patients who would benefit from effective therapy is a key factor in increasing survivability. This work proposes an evolutionary supervised learning method called CancerSig to identify cancer stage-specific microRNA (miRNA) signatures for early cancer predictions. CancerSig established a compact panel of miRNA signatures as potential markers from 4,667 patients with 15 different types of cancers for the cancer stage prediction, and achieved a mean performance: 10-fold cross-validation accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve of 84.27% ± 6.31%, 0.81 ± 0.12, 0.80 ± 0.10, and 0.80 ± 0.06, respectively. The pan-cancer analysis of miRNA signatures suggested that three miRNAs, hsa-let-7i-3p, hsa-miR-362-3p, and hsa-miR-3651, contributed significantly toward stage prediction across 8 cancers, and each of the 67 miRNAs of the panel was a biomarker of stage prediction in more than one cancer. CancerSig may serve as the basis for cancer screening and therapeutic selection..

Brachytherapy can provide sufficient doses to head and neck squamous cell carcinoma (HNSCC) with minimal damage to nearby normal tissues. In this study, the β⁻-emitter ¹⁷⁷Lu was conjugated to DTPA-polyethylene glycol (PEG) decorated gold nanostars (¹⁷⁷Lu-DTPA-pAuNS) used in surface-enhanced Raman scattering and photothermal therapy (PTT). The accumulation and therapeutic efficacy of ¹⁷⁷Lu-DTPA-pAuNS were compared with those of ¹⁷⁷Lu-DTPA on an orthotopic HNSCC tumor model. The SPECT/CT imaging and biodistribution studies showed that ¹⁷⁷Lu-DTPA-pAuNS can be accumulated in the tumor up to 15 days, but ¹⁷⁷Lu-DTPA could not be detected at 24 h after injection. The tumor viability and growth were suppressed by injected ¹⁷⁷Lu-DTPA-pAuNS but not nonconjugated ¹⁷⁷Lu-DTPA, as evaluated by bioluminescent imaging. The radiation-absorbed dose of the normal organ was the highest in the liver (0.33 mSv/MBq) estimated in a 73 kg adult, but that of tumorsphere (0.5 g) was 3.55 mGy/MBq, while intravenous injection of ¹⁷⁷Lu-DTPA-pAuNS resulted in 1.97 mSv/MBq and 0.13 mGy/MBq for liver and tumorsphere, respectively. We also observed further enhancement of tumor-suppressive effects by a combination of ¹⁷⁷Lu-DTPA-pAuNS and PTT compared to ¹⁷⁷Lu-DTPA-pAuNS alone. In conclusion, ¹⁷⁷Lu-DTPA-pAuNS may be considered as a potential radiopharmaceutical agent for HNSCC brachytherapy.

Head and neck carcinoma (HNSC) is often diagnosed at advanced stage, incurring poor patient outcome. Despite of advances in chemoradiation and surgery approaches, limited improvements in survival rates of HNSC have been observed over the last decade. Accumulating evidences have demonstrated the importance of microRNAs (miRNAs) in carcinogenesis. In this context, we sought to identify a miRNA signature associated with the survival time in patients with HNSC. This study proposed a survival estimation method called HNSC-Sig that identified a miRNA signature consists of 25 miRNAs associated with the survival in 133 patients with HNSC. HNSC-Sig achieved 10-fold cross validation a mean correlation coefficient and a mean absolute error of 0.85 ± 0.01 and 0.46 ± 0.02 years, respectively, between actual and estimated survival times. The survival analysis revealed that five miRNAs, hsa-miR-3605-3p, hsa-miR-629-3p, hsa-miR-3127-5p, hsa-miR-497-5p, and hsa-miR-374a-5p, were significantly associated with prognosis in patients with HNSC. Comparing the relative expression difference of top 10 prioritized miRNAs, eight miRNAs, hsa-miR-629-3p, hsa-miR-3127-5p, hsa-miR-221-3p, hsa-miR-501-5p, hsa-miR-491-5p, hsa-miR-149-3p, hsa-miR-3934-5p, and hsa-miR-3170, were significantly expressed between cancer and normal groups. In addition, biological relevance, disease association, and target interactions of the miRNA signature were discussed. Our results suggest that identified miRNA signature have potential to serve as biomarker for diagnosis and clinical practice in HNSC.