Chromosomal instability (CIN) is a characteristic of cancer and a known contributor to sensitivity and resistance to therapy. It refers to the presence of structural and numerical chromosomal abnormalities within tumor cells. While chemotherapy remains an important factor in cancer treatment, it is frequently administered in incorrect ways. Recent research has identified genetic patterns associated with CIN that can classify tumors based on their response to platinum-based chemotherapy.
Researchers validated biomarkers based on CIN signatures to predict treatment resistance in various types of cancer. Using data from 840 individuals in real-world settings, they replicated randomized controlled trials (RCTs) to demonstrate the predictability of therapy resistance. However, the clinical performance and utility evaluation of therapy selection biomarkers are required before their introduction into clinical practice. These assessments should be conducted through a prospective, randomized-controlled biomarker trial.
A single genetic test measures the entire spectrum of signatures in a tumor at the time of diagnosis. To predict platinum resistance, the researchers suggest CIN signatures CX2 and CX3, which are associated with impaired homologous recombination (IHR). Chronic activation of the cGAS-STING pathway leads to a transition of tumors into noncanonical NF-ÎşB signaling, increasing immune evasion and metastasis. The institutional ethics committee approved the prospective Cambridge Translational Cancer Research Ovarian Study 04 (CTCROV04), including the prospective collection of clinical data and samples from patients with high-grade serous ovarian cancer (HGSOC).
The data used in this investigation were collected from the TCGA, along with matching clinical records for 7,105 patients with 33 different types of cancer, and high-quality copy profiles from 7,880 patients. Additionally, an optimal threshold for CX5 activity was determined using 287 paclitaxel-treated cell lines and tumors falling below this threshold were categorized as resistant. Further data used in this study comes from the Hartwig Medical Foundation (HMF), which maintains a multicenter database of 2,979 metastatic cancer patients in the Netherlands.
For sample preparation, 8 ÎĽm sections from formalin and paraffin-embedded tissue blocks were cut. Tumor-enriched regions were retrieved through macrodissection, guided by adjacent hematoxylin and eosin-stained slides. Between 100 ml and 21 volumes of ascitic fluid were collected from patients. Most of the supernatant was extracted from the fluid after gentle centrifugation at 800g for 5 minutes.
Using the Gynecologic Cancer InterGroup criteria, patients were considered eligible if they were 18 years or older, had a recorded tumor stage, progression-free survival (PFS) intervals measured by CA125, and were diagnosed with HGSOC. In case of platinum resistance in primary ovarian cancer, data from 352 patients showed a significantly increased risk of treatment failure (HR=1.46, p=0.0049). In metastatic prostate cancer, 238 patients with taxane resistance exhibited worse outcomes (HR=5.46, p = 0.0003).
Practical routes to clinical translation are suggested by the potential use of these biomarkers in conjunction with liquid biopsies and gene panels. The real-world design of this study was aligned with current regulatory evidence standards, rather than prospective RCT methods. Overall, this work supports the development and clinical validation of CIN signature biomarkers that can predict resistance to multiple chemotherapy classes, including platinum and anthracycline drugs, using a single genetic test.
Reference: Thompson JS, Madrid L, Hernando B, et al. Predicting resistance to chemotherapy using chromosomal instability signatures. Nat Genet. 2025. doi:10.1038/s41588-025-02233-y
