Investigation of a gene signature to predict response to immunomodulatory derivatives for patients with multiple myeloma: an exploratory, retrospective study using microarray datasets from prospective clinical trials
Background Immunomodulatory derivatives (IMiDs), along with proteasome inhibitors, are key components of treatment regimens for multiple myeloma. Nonetheless, outcomes vary among treated individuals. Drug-specific gene-expression profile (GEP) signatures that aid the prediction of favourable and unfavourable outcomes can provide patients with the most effective therapy for their individual disease. We aimed to develop and validate a gene expression signature to suggest which patients would benefit most from IMiD-based therapies. Methods For this exploratory retrospective study, we selected a cohort of patients with newly diagnosed or relapsed or refractory multiple myeloma who were treated in clinical trials with IMiD-containing regimens. Cohorts were eligible if they had publicly available GEP data from patients' bone marrow plasma cells, with long-term follow-up and clinicopathological data. In the development stage of the model, we identified 176 IMiD response genes that were differentially expressed before and after IMiD exposure using pharmacogenomic GEP data from patients who had bone marrow samples taken before and 48 h after a test dose exposure with thalidomide (n=42), lenalidomide (n=18), or pomalidomide (n=18). 14 of these genes had p values less than 0·05 for associations with progression-free survival in patients who received thalidomide in induction and maintenance therapy in the Total Therapy (TT) 2 trial (ie, the training cohort). We combined the 14 genes to create a continuous IMiD-14 score and an optimal cutoff. The subgroup with an IMiD-14 score higher than the cutoff was deemed to be IMiD-resistant. We obtained validation cohorts from four studies of IMiD combination regimens: the TT3a trial (thalidomide in induction and maintenance), the TT3b trial (thalidomide in induction and lenalidomide in maintenance), the TT6 trial (thalidomide in induction and lenalidomide in maintenance), and the vincristine, doxorubicin, and dexamethasone (VAD) group of the HOVON65/GMMG-HD4 trial (thalidomide in maintenance). The primary endpoint was to show the prognostic value of the IMiD-14 gene signature for progression-free survival. Findings In the training cohort, progression-free survival was significantly shorter in the 83 patients with IMiD-14 high scores than in the 92 patients with IMiD-14 low scores; 3 year progression-free survival was 52% (95% CI 42–64) for the IMiD-14 high group versus 85% (78–92) for the IMiD-14 low group, with a hazard ratio (HR) of 2·51 (95% CI 1·72–3·66; p<0·0001). These findings were supported by the results in the validation cohorts, TT3a (115 patients with IMiD-14 high vs 160 patients with IMiD-14 low; 3 year progression-free survival 63% [95% CI 55–73] vs 87% [82–92]; HR 1·54 [1·11–2·15], p=0·010), TT3b (77 patients with IMiD-14 high vs 89 patients with IMiD-14 low; 62% [52–74] vs 80% [72–89]; HR 2·07 [1·28–3·34], p=0·0024), TT6 (20 patients with IMiD-14 high vs 36 patients with IMiD-14 low; 39% [22–68] vs 74% [61–90]; HR 2·40 [1·09–5·30], p=0·026), and the VAD group of HOVON65/GMMG-HD4 (65 patients with IMiD-14 high vs 77 patients with IMiD-14 low; 16% [9–28] vs 54% [44–67]; HR 2·29 [1·52–3·45], p<0·0001). Interpretation Our results suggest that the IMiD-14 model has prognostic value in patients with multiple myeloma who are treated with IMiDs. Some genes in the model could provide novel targets for IMiD resistance and therapeutic intervention. The IMiD-14 model warrants evaluation in prospective studies. Funding Conquer Cancer Foundation ASCO Young Investigator Award and the Carolinas Myeloma Research Fund.
|Persistent URL||dx.doi.org/10.1016/S2352-3026(17)30143-6, hdl.handle.net/1765/101723|
|Journal||The Lancet Haematology|
Bhutani, M. (Manisha), Zhang, Q. (Qing), Friend, R. (Reed), Voorhees, P.M, Druhan, L.J. (Lawrence J), Barlogie, B, … Usmani, S.M. (2017). Investigation of a gene signature to predict response to immunomodulatory derivatives for patients with multiple myeloma: an exploratory, retrospective study using microarray datasets from prospective clinical trials. The Lancet Haematology, 4(9), e443–e451. doi:10.1016/S2352-3026(17)30143-6