Part 1 of 3
Click here to view Part 2
Perhaps the hottest concept in CML research today is that of the “deep molecular response”. So it’s worthwhile to examine how this idea emerged and what it means to people living with CML.
In most cases, CML is caused by an acquired genetic mutation called BCR-ABL in the bone marrow. This gene produces proteins (called transcripts) that signal to white blood cells (WBCs) to proliferate, and it is this overproduction of WBCs that is the hallmark of leukemia. The TKI (tyrosine kinase inhibitor) medications used to treat CML have the ability to suppress the production of BCR-ABL transcripts, thereby normalizing the production of WBCs.
When TKIs were introduced a decade ago, the key benchmark for determining a response to treatment was a complete cytogenetic response (CCyR). This involved examining a small number of cells to see if the shortened Philadelphia chromosome (where the BCR-ABL gene is located) was present. CCyR meant that no cells with the defect could be seen under the microscope. The limitation of this approach was that few cells were examined, so it was very likely that leukemia was hiding elsewhere in the bone marrow.
A more sensitive test was PCR (polymerase chain reaction), which can detect leukemia at the molecular level (i.e. by identifying the presence of the transcripts themselves). According to this measure, a “molecular response” was any reduction in the transcript number. A 1-log reduction meant that the number of transcripts was reduced to 10% of what it had been before treatment started (compared to an international standard rather than the person’s own starting point). A 2-log reduction meant that the number of transcripts was reduced to 1% of what it had been. This 2-log reduction roughly corresponded to CCyR.
At first, PCR offered two main advantages over cytogenetic testing. It could be performed on blood samples, so it was easier (and less painful) to do. And it was a more sensitive measure for detecting very low levels of leukemia disease activity.
A third advantage soon became apparent. If a person could achieve a 3-log reduction (i.e. a reduction to 0.1%), this corresponded to a much lower risk of progressing to accelerated-phase or blast-crisis CML (Hughes and colleagues. N Engl J Med 2003;349:1423-1432; Press and colleagues. Blood 2006;107:4250-4256). As a result, a 3-log reduction – called a major molecular response (MMR) – became the new goal of TKI treatment.
This new benchmark of treatment response is reflected in treatment guidelines produced by the LeukemiaNet group. The goals of treatment they outlined are to achieve a 1-log reduction within the first three months of treatment, a 2-log reduction within the first six months, and a 3-log reduction within 12 months (Baccarani and colleagues. Blood 2013;122:872-884; free full text at www.bloodjournal.org/content/bloodjournal/122/6/872.full.pdf). Treatment goals are less stringent according to the criteria developed by the U.S. National Comprehensive Cancer Network (NCCN). They recommend a 1-log reduction at three and six months, and CCyR (i.e. a 2-log reduction) at 12 and 18 months (Booklet available at www.nccn.org/patients/guidelines/cml/index.html#60).
An important idea in both of these guidelines is that suppression of leukemia activity isn’t the only consideration. Time matters – because as leukemia cells are slowly killed off, there’s an opportunity for other drug-resistant forms to emerge. The situation is akin to having a bacterial infection: you want the antibiotic to eliminate the bug before it can cause more damage or develop resistance to treatment. So early, robust suppression of leukemia is best.
The importance of early suppression was shown in two studies. The German CML Study Group looked at over 1,300 people starting treatment with a TKI (Hanfstein and colleagues. Leukemia 2012;26:2096-2102). The estimated overall survival at 5 years for people who achieved a 1-log reduction in the first three months was 94%. If their response was better (a 2-log reduction at three months), the estimated survival was 97%. But if people didn’t have this degree of early leukemia suppression, the estimated 5-year survival dropped to 87%. Researchers in the UK reported similar findings. If a 1-log reduction was not achieved at three months, the estimated 8-year survival was 57%, compared to 93% if this milestone was reached (Marin and colleagues. J Clin Oncol 2012;30:232-238).
These studies set the stage for the next questions for researchers to examine. If a 3-log reduction is good, would a 5-log reduction be better? And does a 5-log reduction mean that CML is cured?