Investigating biomarkers in clinical samples that are predictive of response in multiple myeloma
Rajpal, Rajesh (2012) Investigating biomarkers in clinical samples that are predictive of response in multiple myeloma. PhD thesis, Dublin City University.
Full text available as:
Multiple Myeloma is a heterogeneous group of disorders of plasma cell proliferation both genotypically and phenotypically. Thalidomide is an effective treatment for multiple myeloma. However, some patients fail to respond to thalidomide treatment and identifying this cohort may provide better individualized treatment. Proteomic analysis was performed on serum samples collected prior to treatment from 39 newly-diagnosed multiple myeloma patients receiving thalidomide-based regimens (22 responders; 17 non-responders). Serum samples were initially immunodepleted to enrich for low-abundance proteins, followed by 2D-DIGE separation and subsequent mass spectrometry to identify differentially expressed protein spots. ELISA-based assays were used to validate the candidate protein biomarkers using unfractionated serum samples from 51 consecutive newly-diagnosed multiple myeloma patients (29 responders; 22 non-responders). Six serum proteins exhibited a statistically significant difference in abundance level between the thalidomide responders and non-responders. Vitamin D binding Precursor (VDB), Transthyretin (TYR), Zinc alpha 2-glycoprotein (ZAG), Serum Amyloid A protein (SAA), Beta-2-microglobulin (B2M), had higher abundance level with fold changes of 1.31 (p=0.00044), 1.32 (p=0.0077), 1.48 (p= 0.0000022), 3.01 (p=0.006) and 1.96 (p=0.0015) fold increase in non-responders compared to responders, respectively. In contrast Haptoglobin (Hp) had a lower abundance level in non-responders compared to responders with a fold change of 3.01 (p=0.0015). Logistic Regression (LR) models were constructed and receiver operating characteristic (ROC) curve analyses carried out on all possible combinations of the differentially expressed proteins. Using logistic regression models, the best possible area under the curve (AUC) was 0.96 using ZAG, VDB and SAA in combination. A more stringent Leave-one-out-cross-validation (LOOCV) indicated an overall predictive accuracy of 84% with associated sensitivity and specificity values of 81.8% and 86.2% respectively. Subsequently 16 of 22 thalidomide-refractory patients successfully achieved complete response or very good partial response using second-line treatment suggesting that the biomarker profile is specific to thalidomide response rather than identifying multiple myeloma patients refractory to all therapies. Using this novel panel of predictive biomarkers, the feasibility of predicting response to thalidomide-based therapy in previously untreated MM has been demonstrated. This approach has the potential for individualizing therapy for patients presenting with newly diagnosed MM.
Archive Staff Only: edit this record