PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Chemotherapy response prediction”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources
PubMed2026

Whole-Genome Deep Learning Predicts Chemotherapy Response in Colorectal Cancer.

Chemotherapy response in colorectal cancer (CRC) exhibits significant heterogeneity, with current clinical predictors failing to capture complex genomic determinants of resistance. We developed a hybrid deep learning framework integrating convolutional neural networks (CNNs) and bidirectional long short-term memory (BiLSTM) networks to analyze whole-genome somatic mutations, evolutionary conservation, chromatin accessibility, and 3D genome architecture in 2,546 TCGA patients. An attention mechanism identified predictive genomic regions. The model achieved an AUC of 0.92 (95% CI: 0.89-0.94) in cross-validation and 0.88 (95% CI: 0.85-0.91) in independent validation, outperforming clinical models (&#x394;AUC = +0.18, p < 0.001). Key predictors included non-coding variants in TP53, KRAS, and PIK3CA regulatory regions. Triple-positive patients (mutations in all 3 regions) had significantly worse progression-free survival (HR = 4.7, p < 0.001). Our framework enables accurate chemotherapy response prediction and reveals novel non-coding resistance mechanisms, advancing precision oncology in CRC.

Humans↗
PubMed2026

Patient-derived organoids predict responses to chemotherapy and PARP inhibitors in advanced ovarian cancer.

BACKGROUND: While tumor organoids hold promise for personalized medicine, clinical validation of epithelial ovarian cancer (EOC) organoids as predictors of therapeutic efficacy-particularly for PARP inhibitors (PARPi)-remains unestablished. METHODS: Patient-derived organoids (PDOs) were established from treatment-naive EOC specimens and characterized by H&E staining, immunohistochemistry, and whole-exome sequencing. Drug sensitivity testing (DST) was performed using carboplatin, paclitaxel, and PARPi (olaparib and niraparib). Clinical homologous recombination deficiency (HRD) status was assessed by tumor sequencing. Organoid responses were prospectively compared to patient outcomes after first-line chemotherapy (carboplatin/paclitaxel) and PARPi maintenance. RESULTS: PDOs were successfully established from 21 of 30 patients (70%) across multiple EOC subtypes and preserved the histopathological features and genomic landscapes of their corresponding primary tumors. Organoid-based DST accurately predicted responses to first-line carboplatin/paclitaxel, with a sensitivity of 100% (95% CI 62.88-100%), specificity of 66.67% (95% CI 12.53-98.23%), accuracy of 91.67% (95% CI 61.52-99.79%), AUC of 0.95 (95% CI 0.85-1.00), and Cohen's kappa of 0.75 (95% CI 0.30-1.00). In evaluating PARPi response, organoids revealed discrepancies between genomic HRD status and actual drug responses. One HRD-positive PDO was PARPi-resistant, consistent with patient non-response, while two HRR-proficient PDOs showed PARPi sensitivity and corresponding clinical benefit. CONCLUSIONS: EOC-derived PDOs provide a robust platform for predicting chemotherapy response and offer added value in assessing PARPi efficacy beyond genomic profiling. Combination of organoid-based testing with genomic analysis may improve precision treatment strategies in EOC.

Humans↗
PubMed2025

Biomarker-Based Nomogram to Predict Neoadjuvant Chemotherapy Response in Muscle-Invasive Bladder Cancer.

Background/Objectives: The aim of this study was to identify response prediction and prognostic biomarkers in muscle-invasive bladder cancer (MIBC) patients undergoing neoadjuvant chemotherapy (NAC). Methods: A retrospective multicentre study including 191 patients with MIBC who received NAC previous to radical cystectomy (RC) between 1996 and 2013. Gene expression patterns were analysed in 34 samples from transurethral resection of the bladder (TURB) using Illumina microarrays. The expression levels of 45 selected differentially expressed genes between responders and non-responders to NAC were validated by quantitative PCR in an independent cohort of 157 patients. Regression analysis was used to identify predictors of downstaging and relapse. A nomogram for predicting downstaging and relapse-including clinicopathological and gene expression variables-was developed. Results: The expression levels of 1352 transcripts differed between responders and non-responders to NAC. A nomogram based on the most predictive clinical variables (age, Tis (in situ), gender, history of NMIBC, and lymphadenopathy) and genes selected following the Akaike information criterion (AIC) (CBTB16, CHMP6, DDX54, CASP8, LOR, and PLEC) was then created. In addition, a three-gene expression prognostic model to predict tumour relapse was generated. This model was able to discriminate between two groups of patients with a significantly different probability of tumour relapse (HR: 2.11; CI: 1.16-3.83, p = 0.01). Conclusions: Our nomogram based on gene expression and clinical data is a useful tool to predict downstaging and tumour relapse after NAC in MIBC patients. Further validation is warranted.

bladder cancer↗
PubMed2026

Single-cell profiling reveals a novel CAF subpopulation linking stromal heterogeneity to immune suppression in breast cancer subtypes.

BACKGROUND: The tumor microenvironment critically influences breast cancer (BC) progression, immune surveillance, and therapeutic response. Cancer-associated fibroblasts (CAFs), a heterogeneous stromal population, are key regulators of these processes, yet their subtype-specific contributions in BC remain insufficiently defined. METHODS: We integrated three single-cell RNA sequencing datasets from 29 BC patients to characterize stromal populations. Bulk RNA-seq data from The Cancer Genome Atlas (TCGA) were analyzed to assess correlations between CAF subsets and immune infiltration. Gene signatures were derived to identify subtype-specific CAF-immune interactions, prognostic markers, and potential predictors of chemotherapy response. RESULTS: Three conserved stromal populations (iCAFs, myCAFs, and pericytes) were identified, along with a previously unrecognized subset, the cluster 3 (CL3) CAF-like cells, referred as metabolic stressed CAF (msCAF). msCAF cells displayed transcriptional programs associated with antigen presentation, stress response, glycolysis, and extracellular matrix remodeling. Their abundance was inversely correlated with T-cell infiltration and function, in a subtype-specific manner: triple negative breast cancer (TNBC) was enriched for msCAFs in immune-infiltrated but functionally constrained microenvironments, whereas Luminal A tumors exhibited weaker immune infiltration with heterogeneous CAF-immune associations. msCAFs were characterized by a conserved gene signature (HLA-A, HLA-C, IL32, EMP3) and subtype-specific genes related to T-cell exhaustion. Several genes demonstrated prognostic relevance with distinct patterns in Luminal A (IER3, TIMP1, TBX3, SEC61G) and TNBC (ADM, C4orf3, LDHA) tumors, as well as shared biomarkers (FN1, LOXL2, P4HA1). Multiple msCAF genes also predicted chemotherapy response, suggesting utility as treatment stratification biomarkers. CONCLUSION: msCAFs represent a clinically relevant CAF subset that drives immune suppression, impacts subtype-specific prognosis, and influences therapy response in BC. These findings highlight msCAFs as promising targets for enhancing immunotherapy and personalizing treatment strategies.

Humans↗
PubMed2025

Lactate dehydrogenase a is a crucial biomarker that affects the prognosis, chemotherapy effect, and immune infiltration of breast cancer.

PURPOSE: Lactate dehydrogenase A (LDHA) is a key node in tumor growth, metabolism, and invasion and is upregulated across multiple cancers. However, the molecular mechanisms by which LDHA influences breast cancer (BC) remain unclear. We analyzed public datasets and an institutional cohort to clarify the relationship between LDHA and BC, with the aim of informing future therapeutic strategies. PATIENTS AND METHODS: Using The Cancer Genome Atlas (TCGA), we assessed LDHA expression in BC and examined its associations with tumor mutational burden (TMB), immune cell infiltration, immune checkpoint molecules, and drug sensitivity. We integrated multiple databases and used Kaplan-Meier analyses to evaluate prognostic value. To explore biological functions of LDHA, we performed Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA). We then analyzed BC patients receiving neoadjuvant chemotherapy (NAC) at Harbin Medical University Cancer Hospital to test the relationship between serum lactate dehydrogenase (LDH) and pathological complete response (pCR). Finally, we used National Health and Nutrition Examination Survey (NHANES) data to examine the association between serum LDH and mortality. RESULTS: LDHA was upregulated in BC tissues, and higher expression was significantly associated with worse overall survival (OS), recurrence-free survival (RFS), and distant metastasis-free survival (DMFS). Functional analyses indicated enrichment of metabolic pathways, such as glycolysis. LDHA expression correlated with multiple immune cell populations, suggesting involvement in the tumor immune microenvironment. Lower LDHA expression was associated with greater sensitivity to several chemotherapeutic agents. In our institutional cohort, patients with lower serum LDH were more likely to achieve pCR, and LDH was an independent predictor of pCR. In NHANES, elevated serum LDH was linked to increased mortality risk. CONCLUSION: Our findings suggest that LDHA expression and serum LDH levels are promising prognostic biomarkers for survival and may predict chemotherapy response in BC patients. These results highlight the clinical relevance of LDHA-mediated metabolic pathways. However, as a correlational study, our findings warrant further validation through functional experiments to confirm LDHA's role as a potential therapeutic target.

Humans↗
PubMed2026

Artificial Intelligence for Diagnosis, Risk Stratification, and Prognosis of Neuroblastoma - A Systematic Review and Meta-Analysis.

PURPOSE: To synthesizes evidence on artificial intelligence (AI) performance in neuroblastoma (NB) diagnosis, risk stratification, prognosis, and genomic characterization. MATERIALS AND METHODS: A systematic review and meta-analysis was conducted following PRISMA 2020 guidelines (PROSPERO: CRD42024539475) across five databases. Meta-analyses used random-effects models with logit-transformed Area Under the Curve (AUCs) and cluster-robust standard errors. AI models were classified as Machine Learning Models (MLM) or Hybrid Nomograms (HN) based on their construction methodology. RESULTS: Of 3,742 articles identified, 53 were included. MLMs demonstrated higher point estimates than radiologists in differential diagnosis (AUC: 0.87 vs. 0.83), though this difference was not statistically significant and carried substantial uncertainty. HNs achieved stronger performance in risk stratification (AUC: 0.87). AI-derived nomograms (AUC: 0.9) and gene signatures (AUC: 0.8) outperformed conventional prognostic markers descriptively. Chemotherapy response prediction remained below clinical utility thresholds across all model types. Only 33.9% of models reported calibration and 24.5% underwent external validation. CONCLUSIONS: AI demonstrates proof-of-concept across multiple NB clinical domains. However, clinical adoption remains premature given persistent gaps in external validation, calibration, dataset size, and pediatric-specific model development. Future studies should test these models prospectively in multicenter pediatric cohorts, ideally through COG or SIOPEN, using shared definitions for diagnosis, risk group, treatment response, and survival outcomes.

Humans↗
PubMed1979

The dilemma of estrogen receptors and the response rate to cytotoxic chemotherapy. A problem of comparability analysis.

The question of whether estrogen receptor assay predicts for chemotherapy response and, if so, how, is currently controversial. Two papers have recently appeared with widely differing results. Both papers use retrospective analysis and are so heterogeneous as regards assay criteria for positivity, prognostic variable and drugs used that they cannot in truth be meaningfully compared. Resolution of the question will probably have to await prospectively designed trials.

Antineoplastic Agents↗
PubMed1979

[Prognostic factors in the drug therapy of metastasizing breast cancer].

175 patients with metastatic breast cancer, treated with chemotherapy, were analyzed retrospectively to identify the characteristics of prognostic importance in predicting response to chemotherapy and survival from onset of the chemotherapy. The most significant factors were the sites of metastatic disease and an estimate of the total extent of disease.

Antineoplastic Agents↗
PubMed1979

Prognostic factors in metastatic breast cancer treated with combination chemotherapy.

Six hundred nineteen patients with metastatic breast cancer, treated with a combination of 5-fluorouracil, Adriamycin, and cyclophosphamide, or close variations of this program, with or without immunotherapy were analyzed retrospectively to identify those host, tumor, or treatment characteristics that might be of prognostic importance in predicting response to chemotherapy and survival from onset of the 5-fluorouracil-Adriamycin-cyclophosphamide treatments. Primary tumor characteristics such as size of primary, number of axillary nodes involved, stage at diagnosis, and type of surgery used for primary treatment were not found to be of prognostic significance. Host characteristics such as age, menstrual status, or family history of breast cancer were similarly unrelated to outcome. Non-Caucasian patients had a lower response rate and somewhat shorter survival than did Caucasians. Pretreatment weight loss, poor performance status, and abnormal biochemical and hematological values were of adverse prognostic significance. An estimate of total extent of disease based on criteria for rating extent of involvement at 12 potential sites was a much more important prognostic factor related to response and survival than actual sites of involvement or the traditional "dominant site" classification. There was a trend, however, for patients with bone involvement to have a longer survival than did patients with metastases to other organ sites. Shorter survival times were observed among patients exposed to extensive prior radiotherapy and those who failed to respond to prior hormonal treatment. The prognostic variables identified in this paper should be used for the design and comparison of clinical trials in the future.

Adult↗
PubMed1978

Clinical relevance of the histopathological subclassification of diffuse "histiocytic" lymphoma.

Because diffuse "histiocytic" lymphoma, which is an immunologically heterogeneous disease, responds well to chemotherapy in some but not all patients, we attempted to identify the morphologic features that might correlate with its behavior and prognosis. We identified five histopathological categories in 66 patients: one with an excellent prognosis (large, cleaved cell, six of eight patients surviving for two years); two with an intermediate prognosis (large, noncleaved and mixed follicular-center cell, nine of 18 and eight of 17, respectively, surviving for two years); and two with a poor prognosis (blastic and pleomorphic pyroninophilic, one of 10 and two of 13, respectively, surviving for two years). The differences among categories were significant (P less than 0.02) and not dependent on stage (P greater than 0.20). Tumors of follicular-center origin had a better prognosis than tumors of nonfollicular-center origin (P less than 0.01). Differences in survival were due to differences in complete response rate. Morphologic subclassification of diffuse "histiocytic" lymphoma may be useful in predicting response to chemotherapy and survival.

Adult↗
PubMed1979

Pharmacokinetic parameters: potential for and problems with their use as predictors of response to cancer chemotherapeutic agents.

The value of pharmacokinetic parameters of antitumor agents to predict for response to chemotherapy of patients with cancer is unknown. Factors which would lead to an expectuation that they would be of value in this regard are the requirement that cells be exposed to a minimum concentration of drug for a minimum period of time to produce cell kill and with certain agents the further requirement that this exposure be during a specific critical phase of the cell cycle. The potential limitations of these parameters for response prediction include the fact that they are whole body parameters which do not distinguish between targets for drug response and targets for dose limiting drug toxicity and that the way drugs are usually given clinically is such as to minimize the effect of patient variation in these parameters. Some practical limitations of their use are the need for very sensitive assays for their accurate definition, the need for computer analysis and the dependence of parameter estimates on the use of the correct model. It is concluded that by themselves pharmacokinetic parameters are unlikely to be of predictive value when drugs are used optimally in a clinical setting, but when considered with biochemical and other parameters may contribute to overall response prediction.

Antineoplastic Agents↗
PubMed2026

Feasibility, reliability, and clinical value of genomic assay on pre-therapeutic biopsy for endocrine receptor-positive HER2-negative early breast cancer.

Endocrine receptor-positive (ER+) and HER2-negative breast cancer (BC) represents approximately 80% of all BCs. Most patients are treated with upfront surgery; however, 15%-30% will develop late recurrences. Genomic assay indication is usually based on postoperative pathological data including histology subtype, tumor size, lymph node status, SBR grade, and Ki67. Performing genomic testing on core needle biopsy specimens prior to surgery may offer several advantages. In this manuscript, we assess the feasibility, reliability, utility, and potential benefits of such genomic analyses performed on core needle biopsies. Several factors may lead to proposing genomic assay analysis on biopsy: (1) optimization of the patient pathway by reducing time to therapeutic decision-making, (2) predicting response to neoadjuvant chemotherapy (NAC) or neoadjuvant endocrine therapy (NET), and (3) refining prognostic assessment to guide adjuvant chemotherapy decisions in patients for whom axillary surgery is not planned. Given the feasibility and reliability of genomic assay on core needle biopsies, it can be suggested that this practice may become more common in the near future. Knowledge of the evolutive risk determined by the result of genomic assay, as well as clinicopathological characteristics, allows more precise personalization of the therapeutic strategy, including the choice between upfront surgery and neoadjuvant therapy, the selection of systemic treatments, and decision-making in the absence of axillary staging.

breast cancer↗
PubMed2025

Personalizing chemotherapy drug selection using a novel transcriptomic chemogram.

Gene signatures predictive of chemotherapeutic response have the potential to extend the reach of precision medicine by allowing oncologists to optimize treatment for individuals. Most published predictive signatures are only capable of predicting response for individual drugs, but most chemotherapy regimens utilize combinations of different agents. We propose a unified framework, called the chemogram, that uses predictive signatures to rank the relative predicted sensitivity of different drugs for individual tumors. Using this approach, providers could efficiently screen against many therapeutics to optimize chemotherapy at any time, whether it be for a treatment-naive tumor or a chemo-resistant tumor requiring a new treatment strategy. To demonstrate the utility of the chemogram, we used predictive signatures (extracted from a previously established method) in our framework to rank predicted sensitivity among drugs within cell lines. We then compared the rank order of predicted and observed response against each drug. Across most cancer types, chemogram-generated predictions were more accurate than predictions made by randomly generated gene signatures, signatures extracted from differential expression alone, and was comparable to another established method of drug response prediction. Our framework demonstrates the ability of transcriptomic signatures to not only predict chemotherapeutic response, but also correctly assign rankings of drug sensitivity on an individual basis. Additionally, scaling the chemogram to include more drugs does not compromise accuracy.

Humans↗
PubMed1979

Clinical utility of lymphocyte surface markers combined with the Lukes-Collins histologic classification in adult lymphoma.

To determine whether analysis of lymphocyte surface markers adds clinically useful information to the Lukes-Collins classification of lymphomas, tumors from 107 adults were histologically classified and studied for surface markers. Ninety-six cases were histologically classified as Lukes-Collins B-cell lymphomas; 87 showed B and one showed T surface markers, whereas eight had neither marker. Eleven lymphomas were histologically T-cell tumors; four of the 11 showed T surface markers, and seven had neither marker. Both the Lukes-Collins classification and surface markers identified patient groups with different clinical characteristics, chemotherapeutic responsiveness and survival. However, by combining surface markers and histologic features, additional important therapeutic and prognostic information was obtained. In each histologic class, patients whose lymphomas failed to express immunologically the histologically predicted marker had fewer responses to chemotherapy and shorter survivals than patients whose lymphomas expressed the predicted marker. Our data suggest that the analysis of surface markers in combination with the Lukes-Collins classification identifies many patients who respond poorly to current therapy and who thus require new therapeutic approaches.

Aged↗
PubMed2025

Integration of Gene Expression and Digital Histology to Predict Treatment-Specific Responses in Breast Cancer.

Deep learning models applied to digital histology can predict gene expression signatures (GES) and offer a low-cost, rapidly available alternative to molecular testing at the time of diagnosis. We optimized transformer-based models to infer GES results and applied this approach to pre-treatment H&E-stained biopsies from 1,940 breast cancer patients treated with neoadjuvant chemotherapy in clinical trial and real-world cohorts. The most predictive histology-derived GES for pathologic complete response (pCR) in the I-SPY2 trial was validated in four external cohorts: CALGB 40601, CALGB 40603, a trial of durvalumab plus CT, and standard-of-care CT-treated patients from the University of Chicago. Among HER2-negative patients, a transformer-based model trained using a signature composed of estrogen-regulated genes, proliferation, apoptosis, and interferon response genes predicted pCR with an AUC of 0.794, outperforming models based on clinical features alone (AUC 0.704, p = 0.001), pathologist TIL assessment, and a model trained directly to predict response from I-SPY2 cases. Tertiles of this signature stratify patients into clinically relevant groups with increasing likelihood of complete response, with pCR rates &#x2265;50% in the top tertile regardless of treatment or hormone receptor status. Additional transformer-based signature models predicted response to specific therapies (but not chemotherapy alone), including a HER2 signaling signature in IO-treated patients, and a claudin-low signature in bevacizumab treated patients. In HER2- cohorts with available gene expression data and histology, models trained on expression data performed similarly to digital histology predictions, but the combination of gene expression and histology outperformed histology alone. These findings suggest that histology-based GES provides additive information to RNA sequencing data and can inform precision treatment selection across breast cancer subtypes.

Journal Article↗