Powered By Bing

Supplementary Information

Gene-expression patterns in drug resistant acute lymphoblastic leukemia cells and response to treatment.

Supplemental Table 10 : Association of the drug resistance gene expression scores and minimal residual disease (MRD).

To test the relation between the combined drug resistance gene expression score (GE score) and the presence (=0.01 percent) or absence (=0.01 percent) of minimal residual disease (MRD) on day 19 of treatment and on day 43 (end of induction), we used MRD data that were available for a subset of the SJCRH cohort (independent test set). MRD results were available on 33 patients on day 19 and 44 patients on day 43. Despite the small number of patients for whom both MRD and gene expression were available in our cohort, there was an association between the GE score and MRD on day 19 (a), with a resistant (high) GE score associated with presence of MRD  (P=0.10, Wicoxon rank sum test). In contrast, there was no evidence of an association between the GE score and MRD on day 43 (b) (P=0.52, Wicoxon rank sum test). The predominant components of treatment were prednisolone, vincristine, asparaginase and daunorubicin (PVAD) during the first 19 days, whereas etoposide and cytarabine were also given between days 19 and 43, consistent with the GE score being an important predictor of disease response to these four agents (i.e., day 19 MRD). Over the entire course of treatment (2.5 to 3 years) PVAD constituted a substantially greater proportion of treatment than other agents such as etoposide and cytarabine, consistent with GE score being significantly related to long term outcome, as observed in the training set and the independent test set.

a. MRD day19
  Combined drug resistance gene expression core Total
  Low (<4.7) Intermediate
(4.7-5.58)
High (>5.58)
MRD day 19
≤0.01%
5 11 2 18
MRD day 19
≥0.01%
2 7 6 15
Total 7 18 8 33
b. MRD day43
  Combined drug resistance gene expression score Total
  Low (<4.7) Intermediate
(4.7-5.58)
High (>5.58)
MRD day 43
≤0.01%
7 18 4 29
MRD day 43
≥0.01%
4 7 4 15
Total 11 25 8 44

for prednisolone, vincristine, asparaginase, daunorubicin

When MRD is included in a multivariate survival analysis with the GE score (n=33), there is an association of a poor outcome in patients with a resistant (high) GE score, after adjusting for MRD status, but the sample size limits the power of such an analysis in our patient cohorts. (a) Day 19 MRD negative status (≤0.01 percent) was associated with a lower probability of relapse (hazard ratio of 0.014, P=0.09), and a drug resistant (high) GE score was associated with a higher risk of relapse (hazard ratio = 2.69, P=0.13), after adjusting for MRD. (b) Using day 43 MRD data (N=44), the hazard ratio for MRD negative patients was 0.04 (P=0.001), and for patients with a resistant (high) GE score the hazard ratio was 2.38 (P=0.08) after adjusting for MRD status.

a.
Variable No. of patients Hazard Ratio  (95% C.I.) P-value
Combined drug resistance gene expression score 33 2.69 (0.75, 9.36) 0.13
MRD day19
MRD day19 ≥0.01% 15 1.0*  
MRD day19 ≤0.01% 18 0.14 (0.014, 1.34) 0.09
b.
Variable No. of patients Hazard Ratio  (95% C.I.) P-value
Combined drug resistance gene expression score 44 2.38 (0.88, 6.44) 0.08
MRD day43
MRD day43 ≥0.01% 15 1.0*  
MRD day43 ≤0.01% 29 0.04 (0.006, 0.30) 0.001

*Reference subgroup
for prednisolone, vincristine, asparaginase, daunorubicin

It is also important to note that even if MRD were to overwhelm the prognostic significance of the gene expression (GE) score, this would not obviate the importance of the association between gene expression and drug resistance or disease outcome. The primary reason is that MRD is a marker of poor response, but it tells one nothing about why a given patient has had a poor response. The gene expression signature provides insights into the basis of drug resistance and poor response, thereby offering strategies whereby more effective therapy can be pursued

< Previous | Next >