Introduction Accurate identification of epidermal growth factor receptor mutation status is essential for treatment planning in locally advanced non-small cell lung cancer, but tissue testing may be limited by specimen availability and sampling. We developed and externally validated a fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography habitat radiomics model to estimate mutation probability as an adjunct to molecular testing. Methods This retrospective multicenter study included 313 patients with locally advanced non-small cell lung cancer who underwent imaging and tissue-based mutation testing. Center A provided the training cohort (n = 142) and internal validation cohort (n = 61), and Center B provided the external validation cohort (n = 110). Separate Otsu thresholds were applied to positron emission tomography and computed tomography intensities within each segmented tumor. Voxel-wise intersection generated four mutually exclusive habitats. Habitat features were selected in the training cohort and evaluated with six classifiers. Additional analyses examined adenocarcinoma-restricted performance, alternative partitioning, segmentation reproducibility, calibration, decision-curve performance, and selection bias. Results The training, internal validation, and external validation cohorts included 69/73, 31/30, and 64/46 mutation-positive/mutation-negative tumors, respectively. The support vector machine habitat model achieved areas under the receiver operating characteristic curve of 0.911 (95% confidence interval, 0.869-0.948), 0.898 (0.819-0.967), and 0.884 (0.829-0.933), respectively. With the threshold selected only in the training cohort, external validation sensitivity, specificity, accuracy, and negative predictive value were 0.750, 0.739, 0.745, and 0.680. The combined model achieved an external validation area under the curve of 0.904 (0.855-0.949). Conclusion Habitat radiomics showed promising discrimination across two centers, including among patients with adenocarcinoma. The model may help prioritize confirmatory testing but should not replace biopsy, plasma testing, or molecular analysis. Prospective validation and direct biological correlation are required before clinical implementation.
Introduction In head and neck cancer (HNC), the prognostic value of [ 18 F]FDG PET/CT SUVmax is limited. Volumetric parameters such as metabolic tumor volume (MTV) and lean body mass-normalized SUV (SUL) may improve risk stratification, particularly in patients with higher body mass index. Methods We conducted a retrospective cohort study of 28 patients with histologically confirmed non-metastatic squamous cell HNC undergoing pretreatment [ 18 F]FDG PET/CT and definitive (chemo)radiotherapy. Primary tumor (P), neck nodal (N), and total (T) MTV, SUVmax, and SULmax were measured. Associations with 2.5-year overall survival (OS2.5), local (LOC), regional (REG), and distant (M+) failure were assessed using logistic regression, receiver operating characteristic (ROC) analysis, and Kaplan–Meier curves. Statistical analyses were performed in R (v3.5); p < 0.05 considered statistically significant. Results The median age was 55 years (IQR, 45–65), and most patients were female (78.6%). Primary tumor sites were nasopharynx (57.1%) or oropharynx (39.3%); one patient had an unknown primary tumor. All tumors were squamous cell carcinomas. Median follow-up was 37.6 months and 11 patients (39%) died. MTV-T showed the strongest association with OS2.5 (AUC 0.85, threshold 14 ml, p = 0.003), followed by SULmax-N (AUC 0.78, p = 0.025), SUVmax-N, and MTV-P (both AUC 0.77, p = 0.03). For distant failure, SULmax-N and SUVmax-N were the strongest predictors (both AUC 0.90, p = 0.0012), outperforming volumetric parameters. Voxel-based primary tumor uptake metrics did not predict local or regional failure, which were associated with T and N stage and MTV-P. SUL metrics showed high correlation with body weight-normalized SUVs (r=0.99–1.00) but no discriminatory advantage. Conclusion Volumetric PET parameters outperformed voxel-based uptake metrics for overall survival prediction, while nodal SULmax and SUVmax better predicted distant failure. Lean body mass normalization did not improve prognostic performance. These findings support PET-derived volumetric and metabolic intensity parameters for risk stratification and require validation in larger cohorts.
Multiple myeloma (MM) remains incurable despite major therapeutic advances, underscoring the unmet need for earlier and genetically informed intervention. The rationale for such intervention lies in the stepwise evolution of MM from precursor states, including monoclonal gammopathy of undetermined significance (MGUS) and smoldering multiple myeloma (SMM), during which genetic alterations accumulate and drive the emergence, expansion, and evolution of malignant clones. In 2025, three landmark cohort studies using whole-exome or whole-genome sequencing (WES/WGS) proposed distinct but relevant genomic frameworks for reclassifying disease risk in MGUS and SMM. Despite differences in design and analyses, these frameworks converge on several principles: (i) the malignant genomic architecture of myeloma is established well before clinical progression; (ii) a substantial fraction of clinically defined precursor disease already harbors the hallmarks of overt myeloma; and (iii) WES/WGS-derived features meaningfully improve clinical risk models. In this perspective, we critically reviewed these three contemporaneous frameworks and synthesize their convergent implications, not previously consolidated for clinical use, into a practical approach toward genomics-guided management of myeloma precursor conditions.
Introduction Locally advanced rectal cancer receives neoadjuvant chemoradiotherapy (nCRT) as standard clinical management. Reliable non-invasive biomarkers for predicting pathological complete response (pCR) after nCRT remain urgently needed. This study aimed to explore correlations between peripheral blood lymphocyte subsets and pCR rate in rectal cancer patients receiving nCRT, and screen predictive immune markers for individualized therapeutic decision-making. Methods This was a retrospective cohort study. A total of 143 rectal cancer patients receiving nCRT in our center between June 2021 and December 2022 were consecutively enrolled. Tumor regression grade (TRG) was applied to evaluate pathological treatment response. Kolmogorov–Smirnov test, Mann–Whitney U test, chi-squared test, binary logistic regression and ROC curve analyses were performed to investigate associations between lymphocyte subsets and pCR. Results A total of 48 patients (33.6%) achieved pCR after nCRT. Absolute counts of all detected peripheral blood lymphocytes after nCRT. Gender and age were correlated with lymphocyte subsets: males had higher absolute counts and percentages of NK cells than females (P<0.05); patients younger than 65 years had higher percentages of CD3+T cells and CD3+CD4+T cells, and lower NK cells (P<0.05). Multivariate logistic regression demonstrated that age <65 years (P=0.038), TNM stage I (P=0.026), longer interval between radiotherapy and surgery (P=0.003), and higher percentage of CD3+T cells (P=0.043) were independent predictive factors for pCR. ROC analysis revealed that CD3+T-cell percentage yielded an AUC of 0.673 (95% CI:0.581-0.764) for pCR prediction, with optimal cutoff value 68.155, sensitivity 0.563 and specificity 0.726. Conclusion The percentage of peripheral blood CD3+ T cells may serve as a non-invasive predictive marker for the efficacy of neoadjuvant chemoradiotherapy (nCRT) in rectal cancer (RC) patients. Combined with age, tumor stage, and treatment interval, it may help optimize individualized treatment regimens.
Introduction Conventional conformity indices primarily quantify the geometric relationship between the target volume and the prescription isodose volume, which may not fully reflect clinically relevant dose deviations. This study aimed to develop a medical conformity index (MCI) that integrates geometric conformity with clinical considerations and to evaluate its performance in assessing radiotherapy plan quality. Methods The proposed MCI quantifies conformity by incorporating bidirectional local distance between the target volume (TV) and Prescription Isodose Volume (PIV) using an asymmetric Gaussian distance penalty function. Distinct penalty levels are assigned to under- and over-covered regions to reflect the clinical priority of target coverage and organ at risk (OAR) sparing. Additionally, a Target Volume Weight Function (TVWF) is integrated to ensure consistency across different target sizes, and OAR-specific constraints are introduced. The performance of the MCI was evaluated using synthetic 2D plans and clinical treatment plans, compared against conventional indices such as RTOG, SALT-Lomax, and CN. Results The proposed MCI provided more consistent and clinically relevant scoring compared to conventional indices, particularly in cases with complex geometric discrepancies. Unlike existing metrics, the MCI demonstrated superior sensitivity to spatial differences and allowed flexible evaluation through user-defined penalty levels. Furthermore, the index enabled a robust assessment of plan quality by reflecting the clinical impact of dose distribution variations. Conclusions The proposed MCI provides a quantitative conformity measure that integrates bidirectional spatial distance, asymmetric penalties, and target volume weighting. The results demonstrated that MCI can differentiate spatial conformity differences that may not be fully reflected by conventional conformity indices. Further clinical validation is required to determine clinically meaningful penalty levels and acceptance criteria.
