New DNA Tools Crack the Code on Hard-to-Classify Leukemia

Adults can get a type of blood cancer called B-cell precursor acute lymphoblastic leukemia, or B-ALL for short. Doctors know it comes in many different forms. But here is the problem. The usual tests they run cannot figure out what subtype a lot of patients have. About half of the people with a specific kind of this cancer, one that lacks a certain gene change called BCR::ABL1, end up labeled as unclassified. Why does this matter? Because knowing the exact type helps doctors pick the right treatment. Using full-genome or full-transcriptome sequencing on a regular basis is not practical for most hospitals.

Researchers wanted to fix this problem. They tried combining two advanced genetic tools. One is called Optical Genome Mapping, or OGM. The other is called targeted next-generation sequencing, or t-NGS. They tested this combo on 100 adults who had this BCR::ABL1-negative form of B-ALL. All of these patients were part of a study called PETHEMA-LAL-19. The goal was to see if using both tools together could do a better job than the standard tests at sorting patients into the right subtypes.

The results were pretty striking. By using OGM and t-NGS together, doctors could classify 84 percent of the patients into 15 different subtypes. That is a huge jump. The number of patients who stayed unclassified dropped from 50 percent down to just 16 percent. Think about what that means for patients who would have been left in the dark before. Now doctors have a much better chance of knowing what they are dealing with.

Each tool brought something different to the table. OGM was good at spotting a subtype called BCR::ABL1-like, which showed up in 26 percent of cases. People with this subtype were more likely to still have traces of cancer after their first round of treatment. About 72 percent of them had minimal residual disease compared to 34 percent of others. On the flip side, t-NGS was better at finding subtypes defined by specific mutations, like PAX5 P80R and PAX5alt. Patients with these mutations were more likely to be clear of detectable cancer after treatment, with 91 percent showing no minimal residual disease compared to 51 percent of others.

The tools also revealed some interesting patterns. A genetic feature called IKZF1plus was way more common in the BCR::ABL1-like subtype, showing up 62 percent of the time compared to just 12 percent in other cases. JAK-STAT mutations only appeared in patients who had CRLF2 rearrangements. And PAX5-biallelic events were found in 92 percent of PAX5-subtype cases but only 2 percent of others. The researchers found that this combined approach is practical enough to use in everyday diagnostic work.