Pretreatment variables associated with a response to immunosuppression in NIH MDS patients (all ethnicities)
| Favorable factors (n = patients with feature/informative patients) . | Significance in univariate analysis,3-150P . | Significance in multivariate analysis, backward selection,3-151P . | Adjusted odds ratio in logistic regression3-152 . |
|---|---|---|---|
| Positive for HLA-DR15 (n = 29 of 77) | .003 | .008 | 8.53 times more likely to respond if positive for DR15 (95% CI 1.59-45.66) |
| Shorter duration of red cell transfusion dependence (n = 82, continuous variable, mean ± SD = 776 ± 292 d) | .009 | .023 | 1.56 times more likely to respond for each 90-d decrease in duration (95% CI 1.08-2.25) |
| Younger age (n = 82, continuous variable, mean ± SD = 59 ± 2.9 y) | .0002 | .001 | 2.18 times more likely to respond for each 5-y decrease in age (95% CI 1.36-3.48) |
| Less than 5% marrow blasts (n = 60 of 82) | .06 | .834 | |
| Normal cytogenetics (n = 53 of 82) | .077 | .473 | |
| Increased no. of cytopenias (n = 38 of 82) | .033 | .321 | |
| Decreased bone marrow cellularity (n = 20 of 82) | .076 | .145 | |
| Low IPSS score3-153 (n = 68 of 82) | .12 | .120 | |
| PNH present by flow (n = 8 of 77) | .0002 | .136 |
| Favorable factors (n = patients with feature/informative patients) . | Significance in univariate analysis,3-150P . | Significance in multivariate analysis, backward selection,3-151P . | Adjusted odds ratio in logistic regression3-152 . |
|---|---|---|---|
| Positive for HLA-DR15 (n = 29 of 77) | .003 | .008 | 8.53 times more likely to respond if positive for DR15 (95% CI 1.59-45.66) |
| Shorter duration of red cell transfusion dependence (n = 82, continuous variable, mean ± SD = 776 ± 292 d) | .009 | .023 | 1.56 times more likely to respond for each 90-d decrease in duration (95% CI 1.08-2.25) |
| Younger age (n = 82, continuous variable, mean ± SD = 59 ± 2.9 y) | .0002 | .001 | 2.18 times more likely to respond for each 5-y decrease in age (95% CI 1.36-3.48) |
| Less than 5% marrow blasts (n = 60 of 82) | .06 | .834 | |
| Normal cytogenetics (n = 53 of 82) | .077 | .473 | |
| Increased no. of cytopenias (n = 38 of 82) | .033 | .321 | |
| Decreased bone marrow cellularity (n = 20 of 82) | .076 | .145 | |
| Low IPSS score3-153 (n = 68 of 82) | .12 | .120 | |
| PNH present by flow (n = 8 of 77) | .0002 | .136 |
95% CI indicates 95% confidence intervals.
P value was given by the χ2 test in the case of categorical variables and by the Wald test in a logistic regression model for continuous variables.
P values (the last 6 rows) derived from the Wald χ2 test during backward selection for a logistic regression model that initially included all the variables based on 69 informative patients and, also, P values of the first 3 rows from the multivariate analysis.
Adjusted odds ratio derived from logistic regression model including only the first 3 variables based on 74 informative patients.
Low IPSS score is IPSS low risk and intermediate-1.