Acute Lymphoblastic Leukemia

Acute lymphoblastic leukemia (ALL) is the most common childhood cancer, and modern treatment cures more than 85% of children who have it. That success has shifted a real question toward treatment itself: several of the drugs used are hard on the body in ways that vary a lot between patients, and predicting who will react badly, to which drug, has become as practically useful as the original diagnosis.

This page is not about who develops ALL — this site's catalogue has no variant for that. It is about four separate findings on who reacts badly to specific parts of the standard treatment, each from a different study, with genuinely different strength of evidence. This page states that difference rather than presenting all four the same way.

Vincristine and nerve damage — the strongest evidence here

Vincristine is one of the most widely used drugs against childhood ALL, and peripheral neuropathy — nerve damage causing pain, numbness or weakness, usually in the hands and feet — is one of its most common side effects. A study across two clinical trials, 321 patients, found a variant in the promoter of CEP72 (a gene involved in building the cell's internal scaffolding) strongly associated with it. Patients homozygous for the risk allele — 16% of the cohort — developed serious neuropathy 56% of the time, against 21.4% for everyone else, and the neuropathy that did occur was more severe. This is not only a statistical association: lowering CEP72 in human neurons and leukemia cells in the lab made both more sensitive to vincristine, an experimental result pointing the same way as the clinical one.

Asparaginase — an allergic reaction, and a rare cause of pancreatitis

Asparaginase is another core ALL drug, and some children develop hypersensitivity reactions to it that limit how much of the drug they actually receive. Across 3,308 children in five cohorts, a variant in NFATC2 — a gene central to how immune cells activate — was the strongest predictor of that hypersensitivity. Carriers of the variant also had measurably higher NFATC2 expression in their own tumour and reference cell samples, a second, independent line of evidence pointing at the same gene.

A separate, rarer complication is acute pancreatitis, which happens in about 2% of patients on asparaginase. A rare variant in CPA2 — which encodes a pancreatic digestive enzyme — was strongly linked to it in a study of 5,185 patients, with an extremely large estimated effect. That estimate carries a wide margin of uncertainty, exactly what a rare variant produces even with a real underlying effect, and a broader look at other CPA2 variants in the same patients found more of them clustered in those who developed pancreatitis — support for the gene mattering here, even if the single number is imprecise.

A liver-enzyme signal, honestly incomplete

The fourth finding is the least settled. A study of liver enzyme (ALT) elevation after induction therapy found its strongest, best-explained result in PNPLA3 — a gene already well known for fatty liver disease risk in the general population, discovered again here in a cancer-treatment context. That variant is not yet in this site's own catalogue. What is catalogued is a second signal from the same study, near GREB1, which the paper itself does not explain. It is reported here as a real, genome-wide significant finding with no known mechanism, rather than skipped or dressed up with a guess.