Reading Transplant Survival Curves
This National Marrow Donor Program (NMDP) Web site provides outcomes data from the NMDP and from the Center for International Blood and Marrow Transplant Research (CIBMTR). The data from the CIBMTR and the NMDP show information on survival for people who received a bone marrow or cord blood transplant (also called a BMT) to treat their disease. The data include patients with a range of medical situations who underwent various treatment plans for transplant.
The NMDP and CIBMTR data are shown on survival curves (sometimes called Kaplan-Meier curves). Survival curves are useful because you can see patients' survival rate over time. The vertical axis (numbers going up the side) shows the percent of patients who survived. The horizontal axis (numbers along the bottom) shows time measured from the day of transplant. The height of a curve at any point shows the survival rate at that point in time. The number (N) of patients that each curve represents is shown in parentheses, for example (N=116). The shape of the curve shows how patients' risks change over time. A curve that drops steeply at first but then flattens out shows the risks are higher soon after transplant but become lower over time. If a curve slopes gradually downward and does not flatten out, the risks become steadily lower over time.
Often, two or more survival curves are shown on the same figure. This helps you to compare outcomes for different groups, such as patients with a different disease status at the time of transplant. Many NMDP survival curves include a log-rank p-value (such as p < 0.0001). The p-value is a statistical tool to help tell whether differences between the groups shown are real (statistically significant) or might be due to chance. A p-value less than 0.05 is often considered good evidence that differences are real. A p-value of 0.05 means there is a 5% chance — one chance in 20 — that the differences are due to chance. The smaller the p-value, the more likely that the difference between groups is real. For example, a figure of transplant outcomes for patients with leukemia may show one curve for patients in first remission and another for patients in second remission. If the 5-year survival rate of the first group is 20% higher than the rate for the second group, the difference appears large. However, it is important to look at the p-value to find out if this difference is real.
If the p-value is a very small number, such as 0.0001, it is likely that the difference is real. A transplant in first remission had a better chance of 5-year survival than a transplant in second remission. If the p-value is a large number, such as 0.1, there is not enough information to know whether the difference is real. For example, if there are a small number of patients in the study, the difference in outcomes might be caused by factors unique to individual patients.
Talk to your doctor
It is a good idea to ask your doctor for help interpreting these data and any other survival outcomes data you find. Your doctor can help you understand the figures. Your doctor can also explain factors that may affect the data and discuss how they apply to your specific situation.