Short notes and literature interpretations Dr. Sahin shares on X, translated into English. Each note includes the underlying source and a link to the original post.
Could AI help radiologists catch liver cancers that might otherwise be missed? An AI system that analyzes liver CT scans was tested prospectively in 10,333 consecutive patients. The system flagged 51 previously missed lesions, 15 of which were cancer. Ultimately 37 radiology reports were revised and 22 patients were referred for multidisciplinary evaluation. A single-center study whose findings still require radiologist confirmation, but a meaningful step toward real clinical use.
It was wonderful to discuss the opportunities and challenges in precision oncology on the #WINSympo2026 panel today. Thanks to the WIN Consortium for bringing together such a strong group!
A pooled analysis of 1,038 patients with operable melanoma who received treatment before surgery. Among the 735 patients who received immunotherapy, 58% had very little or no viable tumor left at the time of surgery. About 98% of these patients were alive at 5 years.
According to a new study, roughly 2.3 million new cancer cases worldwide in 2024 were attributable to infections. That is approximately 1 in every 8 cancers.
Tumors can use CD47 as a “don’t eat me” signal against macrophages, the immune cells that can swallow cancer cells. In the randomized phase 2 ASPEN-06 trial, adding the CD47 blocker evorpacept to treatment increased tumor response in previously treated HER2-positive gastric and gastroesophageal junction cancers.
A new study shows that some pancreatic cancer cells rely on an interaction with liver cells in order to grow in the liver. Blocking this interaction in mice reduced metastatic growth. An early, preclinical finding, but an intriguing potential treatment strategy.
An important FDA approval in HER2-positive advanced gastric, gastroesophageal junction, and esophageal adenocarcinoma: zanidatamab + chemotherapy + tislelizumab (IHC 3+ or IHC 2+/ISH+) and zanidatamab + chemotherapy (IHC 3+ only). In the phase 3 HERIZON-GEA-01 trial, the triple combination prolonged the time until disease progression (12.4 vs 8.1 months).
In colorectal cancer that has spread to the liver, recurrence remains a significant problem even after surgery. In a randomized trial, chemotherapy delivered directly to the liver (HAI) was compared with standard intravenous chemotherapy. Recurrence in the liver was delayed (25 vs 12 months) and survival showed a trend toward improvement (62% vs 47% at 5 years), but HAI had more side effects. Phase 2 data, but a promising approach that deserves evaluation in phase 3.
Rezatapopt, an oral drug designed to restore function to a specific p53 mutation (TP53 Y220C), showed manageable side effects and early antitumor activity in a phase 1 study in heavily pretreated patients with advanced cancers. A promising step toward mutation-targeted precision therapy.