Chemical Genomics Reveals JAK STAT Activation As a Mechanism of Resistance to HDAC Inhibitors in B Cell Lymphomas
Bibliographic record
Abstract
Abstract Background: HDAC inhibitors (HDACi) are being investigated as treatment for relapsed/refractory non Hodgkin lymphoma (NHL) and other cancers. However, the mechanisms underlying sensitivity and resistance to HDAC inhibition in lymphomas have not been fully characterized. We probed the cellular and molecular response to HDACi in vitro and in vivo in order to determine factors that dictate the response to HDACi and to enable design of approaches to incorporate HDACi into novel combination therapeutics. Methods: High-throughput cytotoxicity screening was performed using two different HDAC inhibitors, LBH589 (panobinostat) and SAHA (vorinostat) in 52 lymphoid cell lines characterized through RNA-seq and microarray gene expression profiling. This screen revealed a greater than 50-fold range in concentration needed to induce cytotoxicity for the 2 different HDAC inhibitors and there was moderate correlation between the 2 compounds in this panel (Pearson correlation r = 0.76, p < 0.01). By pairing this chemosensitivity data with gene expression profiles of the screened cell lines, we developed a gene expression classifier for LBH589 that identified resistant and sensitive cell line groups. This predictor was applied to B-cell NHL cell lines tested with LBH589 in the Cancer Cell Line Encyclopedia (CCLE) and we found that the sensitive and resistant cell line groups distinguished by this method differed more than 5-fold in IC50 (0.021 vs. 1.24 nM, P < 0.01 by Wilcoxon rank sum), thus validating the ability of this approach to distinguish HDACi resistant cell lines. We further initiated a clinical trial of LBH589 in relapsed/refractory diffuse large B cell lymphoma patients combined with RNAseq profiling of their tumors prior to embarking on treatment. We treated nine patients with LBH589, and application of our response predictor to scaled RNAseq gene expression data revealed 4 predicted responders and 5 predicted non-responders. Two of the predicted responders had a clinical response to LBH589, whereas none of the predicted non-responders had a clinical response, thus our classifier was able to identify all of the LBH589-responsive patients from this cohort (P = 0.08 by Fisher's exact test). Analysis of differentially expressed molecular pathways in HDACi sensitive and resistant samples by gene set enrichment revealed the JAK-STAT pathway as the most differentially expressed pathway associated with HDACi resistance (at P < 0.001 and FDR < 0.20). We further identified a number of distinct mutations including STAT3, SOCS1 and JAK1 that were associated with activation of the JAK-STAT pathway by gene expression signatures and the LBH589 response signature in DLBCL cell lines and patient samples by analysis of RNA-seq data. Phosphoprotein analysis by Western blot and Sis-inducible-element (SIE) luciferase reporter assays were used to confirm JAK-STAT activation in these samples and we found that overexpression of STAT3 Src-homology domain mutations activated JAK-STAT3 signaling in isogenic cell lines and fostered resistance to LBH589 in vitro. Conversely, using in vivo DLBCL xenograft models, we found that combining JAK-STAT and HDAC inhibition by treatment with LBH589 and ruxolitinib resulted in synergistic reduction of tumor cell viability and tumor growth with tolerable toxicity in mice. Conclusions: Sustained JAK-STAT activation appears to mediate resistance to HDAC inhibition in DLBCL and other NHLs and several recurrent genetic lesions drive JAK-STAT activation in these diseases. This process can be overcome by JAK 1/2 inhibition with ruxolitinib and these findings demonstrate a role for combination therapy with HDAC inhibitors and small molecules targeting the JAK-STAT pathway in lymphoid malignancies. Disclosures No relevant conflicts of interest to declare.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".