Microsatellite instability (MSI) and tumor mutational burden (TMB) are crucial biomarkers in gastric (GC) and colorectal cancer (CRC), yet their conventional sequencing-based detection is costly and time-consuming. Since only ~20% of patients are MSI-high or TMB-high and likely to benefit from immunotherapy, expensive genomic testing is often unjustified. This study developed a deep learning framework to predict MSI and TMB status directly from routinely available Hematoxylin and Eosin (H&E)-stained whole-slide images, leveraging fused nuclear segmentation features to improve accuracy. Using samples from TCGA (350 GC and 376 CRC for MSI; 400 GC and 387 CRC for TMB), image features were extracted with CLAM and nuclear features with Hover-Net. These features were combined via Multimodal Compact Bilinear Pooling and utilized in six distinct deep learning models. By fusing the nucleus segmentation features, the model increased area under the receiver operating characteristic curve (AUC) by 1%–3% and recall by 5%–11% in five-fold cross-validation, significantly outperforming models that relied solely on image features. External validation on a CRC dataset from the China-Japan Friendship hospital further validated the model's robustness, achieving an AUC of 0.81 and a recall of 0.80 for MSI prediction. Additionally, notable differences in cellular composition were observed across cancer types and clinical groups, emphasizing the pivotal role of cellular features in cancer development. These findings highlight the advantages of integrating H&E-stained image features with nuclear segmentation data and advanced deep learning techniques to improve predictive accuracy and reduce the cost of MSI/TMB testing, potentially advancing personalized cancer treatment strategies.

Tumor-induced osteomalacia is a rare and fascinating paraneoplastic syndrome usually caused by a small, benign phosphaturic mesenchymal tumor. Most tumors are treated surgically, but we are unaware of any reports that compare the results of curettage and segmental resection for lesions in long bones. Seventeen patients (ten male and seven female) with tumor-induced osteomalacia lesions in long bones, who underwent surgical treatment from December 2004 to August 2013 in our hospital, were included in this retrospective study. The mean follow-up (and standard deviation) was 35 ± 27 months (range, twelve to 116 months). The characteristics of the tumor and the effects of different surgical treatments (curettage compared with segmental resection) were evaluated. All patients showed typical clinical characteristics of tumor-induced osteomalacia, including elevated serum fibroblast growth factor-23 (FGF-23); 82% of tumors were in the epiphysis, and 82% grew eccentrically. The mean maximum diameter of the tumors was 2.4 ± 2.0 cm. The complete resection rates were similar for curettage (67%) and segmental resection (80%). However, the recurrence rate after curettage (50%) was higher than that after segmental resection (0%). The complete resection rate for secondary segmental resection (75%) was not different from that for primary segmental resection (83%). All of our cases of tumor-induced osteomalacia were caused by phosphaturic mesenchymal tumors. After successful removal of tumors, serum FGF-23 returned to normal within twenty-four hours and serum phosphorus levels returned to normal at a mean of 6.5 ± 3.5 days. Most lesions in long bones are located in the epiphysis, so curettage is first suggested to maintain joint function. If curettage is incomplete or there is a recurrence, secondary segmental resection should be considered curative. Changes of serum FGF-23 and phosphorus levels before and after the operation may be of prognostic help. Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Pathological examination combined with tumor markers has become a standard for the diagnosis of intracranial germ cell tumors (ICGCTs), but the current concept of ‘secreting germ cell tumors’ and three empirically highly specific diagnostic criteria (β-hCG ≥ 50 IU/L or αFP ≥ 10 ng/mL; β-hCG ≥ 100 IU/L or αFP ≥ 50 ng/mL; β-hCG > 50 IU/L or αFP > 25 ng/mL) are not based upon pathology examination or CSF cytology. Further investigation is needed to re-evaluate their value. A multidisciplinary diagnostic team was created. Valid β-hCG/αFP data were collected from cases of ICGCTs confirmed by pathology and CSF cytology (n = 58) between 1991 and 2012, and from suspected ICGCTs cases (n = 17) between 2011 and 2012 as controls [Langerhans cell histiocytosis (LCH), n = 12; and other intracranial tumor (ICT), n = 5]. The cut-off points for β-hCG and αFP were calculated using receiver operating characteristic (ROC) curves. This study clarifies the relative rationality of one criteria (β-hCG > 50 IU/L and αFP > 25 ng/mL); confirms new β-hCG diagnostic cut-off points: CSF β-hCG ≥ 8.2 IU/L and serum β-hCG ≥ 2.5 IU/L (sensitivity of 47 and 34 %, respectively, specificity of 100 %, both; P < 0.05); and empirically adjusts the criteria for αFP to ≥ 3.8 ng/mL in CSF and to ≥ 25 ng/mL in serum. The total diagnostic sensitivity for ICGCTs finally increased from 34.6 to 65.4 % (P < 0.05, diagnostic value of CSF β-hCG exceeds 90 %). Subtype diagnosis improved with αFP in 16.7 % of non-geminomatous germ cell tumor cases. New evidence-based criteria of β-hCG and αFP can help improving early and formal diagnosis of ICGCTs, and is of great clinical significance.

Adenohypophysis spindle cell oncocytoma (ASCO) is a rare tumor recently reported by Roncaroli et al in 2002. This tumor is considered a grade I tumor by the World Health Organization. We report a rare case of malignant ASCO with repeating recurrences and a high Ki-67 index—a challenging diagnosis guided by clinical presentations, radiological signs, and postoperative pathological tests. We represent a 30-year-old man who had suffered from headaches, diplopia, and impaired visual field and acuity. His magnetic resonance imaging revealed an abnormal sellar mass and was originally misdiagnosed as a pituitary macroadenoma. We present detailed analysis of the patient's disease course and review relevant literature. When surgically treated, the specimen revealed a typical histopathology pattern of ASCO. The tumor recurred for several times and the patient underwent 3 surgeries and 1 γ-knife treatment, which was accompanied by a continuously increasing Ki-67 index. This is the first reported case of malignant ASCO (WHO III–IV grade). Despite its rarity, ASCO should be considered in the differential diagnosis of sellar lesions that mimic pituitary adenomas.