For decades, the first treatment for a fit adult with acute myeloid leukemia (AML) has been intensive induction chemotherapy: typically, 7 days of cytarabine plus 3 days of an anthracycline. The regimen works often enough to remain the standard, but it puts patients in the hospital for nearly a month, drives blood counts low enough to cause serious infections and bleeding, and cures a minority of those with high-risk disease.

Now a phase 2 trial published in September in The New England Journal of Medicine is forcing hematologists to ask whether a gentler regimen can do better.

The PARADIGM trial found that azacitidine-venetoclax, a combination previously reserved for patients too old or frail for chemotherapy, increased the event-free survival by more than twofold than induction chemotherapy in 172 fit adults with newly diagnosed AML.

Patients on the combination spent roughly half as many days in the hospital during the first month and had lower rates of serious infection and bleeding, as Medscape Medical News reported at the time.

Overall survival was similar whether patients received azacitidine-venetoclax or chemotherapy despite the fact that more patients on the combination went on to receive a stem cell transplant (60% vs 40%). An accompanying editorial suggested that some leukemias may resist the immune attack from donor cells after transplant and called azacitidine-venetoclax “unlikely to be curative in most patients with higher-risk disease.”

Longer follow-up is still needed to determine any difference in posttransplant survival between the treatment arms, said Tara Lin, MD, director of the acute leukemia program at the University of Kansas Cancer Center in Kansas City.

But right now, no phase 3 trial is testing the combination in the population studied in the PARADIGM trial.

Despite the unknowns, some centers have already switched their approach. Others say phase 2 data from 172 patients are not enough to abandon a standard that has stood for decades.

Who Has Changed Practice and Who Hasn’t

Amir Fathi, MD, of the Mass General Brigham Cancer Institute in Boston, led PARADIGM, and he told Medscape Medical News that his institution has made azacitidine-venetoclax the default for patients who match the trial’s criteria.

“Those patients who would have in the past gotten intensive induction, now we give a combination of HMA [hypomethylating agents] and venetoclax,” he said.

If those same patients were treated at the OHSU Knight Cancer Institute in Portland, Oregon, they might get either regimen.

Curtis Lachowiez, MD, a hematologist-oncologist at OHSU who studies how leukemia genetics predict response to venetoclax, called azacitidine-venetoclax “one option among several.”

At the University of Kansas, where Lin practices, the first option is a clinical trial. Outside of a trial, she said, for patients with adverse-risk features such as complex chromosomal changes, the frequent choice is an HMA plus a venetoclax-based regimen followed by allogeneic stem cell transplant.

Justin Watts, MD, chief of the Leukemia Section at Sylvester Comprehensive Cancer Center at the University of Miami, landed in between. Watts was first author of an earlier trial that gave azacitidine-venetoclax to adult patients younger than 60 years who had unfavorable-risk disease. He draws his own line at age and risk category.

“Anyone over 50, certainly over 60, who’s fit, with ELN [European LeukemiaNet] adverse risk, you should give them ven-aza [azacitidine-venetoclax] over 7 + 3,” Watts said, referring to the ELN classification and the standard induction regimen. (His recommendation excludes patients with monocytic disease, however.)

“If the patient was unfit, you’d give it, and you’d expect the response rate to be high,” Watts added. “So if a patient was fit, why wouldn’t you give it?”

When Chemo May Still Be Preferred

PARADIGM excluded three groups of patients with established reasons to receive intensive chemotherapy or targeted drugs: those with FLT3 mutations, those with core binding factor AML, and those younger than 60 years with NPM1 mutations.

Because of that, Fathi said he believes the trial data do not apply to those patients.

Among the patients studied, most had intermediate- or adverse-risk disease. But even within that group, leukemia’s biology may steer some patients toward chemotherapy.

Watts’s trial stopped enrolling patients with monocytic AML (a subtype in which leukemia cells resemble monocytes) after an interim analysis found that half of those patients failed to respond to azacitidine-venetoclax. (PARADIGM did not exclude patients with monocytic disease.)

Watts and his co-authors concluded that patients with monocytic disease who can tolerate intensive chemotherapy should receive it.

Lachowiez pointed to mutations in growth-signaling genes such as RAS as another reason to favor chemotherapy, even in patients who meet PARADIGM’s criteria.

“I don’t know that we can say quite confidently yet that anybody that fits the inclusion criteria should absolutely get aza-ven,” he said. “I think that would be a little too strong of a statement.”

Can We Predict Who Benefits?

Clinicians choosing between azacitidine-venetoclax and intensive chemotherapy do not yet have a validated tool to guide the decision. Lachowiez and his colleagues have developed the PRISM risk model, which uses 17 clinical and genetic variables to predict survival after venetoclax-based treatment.

But PRISM was built to estimate prognosis on a given regimen, not to compare two regimens head-to-head.

“PRISM’s use is really restricted to patients receiving HMA plus venetoclax and not necessarily used to decide between those two approaches right now,” Lachowiez said.

Another unresolved issue is whether azacitidine-venetoclax clears disease as deeply as chemotherapy does. In PARADIGM, flow cytometry found similar rates of undetectable residual leukemia in both groups, but a commentary in The ASCO Post argued that current tests may not be sensitive enough to prove that a gentler regimen eliminates as much leukemia as intensive chemotherapy does.

The PARADIGM investigators are now analyzing samples with DNA sequencing, a more sensitive measure. Fathi said sequencing results are expected within a year or two.

What Comes Next?

Typically, a confirmatory phase 3 trial in the same population is necessary for treatment guidelines to change. But Fathi said a larger trial was beyond the scope of what an investigator-initiated study could accomplish.

Several randomized trials with related designs are enrolling internationally, but none replicate the PARADIGM population. One, myeloMATCH, a National Cancer Institute-funded trial in the US, is testing azacitidine-venetoclax and other regimens in younger patients with adverse-risk disease. The trial has enrolled 30% of a planned group of 335 patients and is projected to complete enrollment in late 2027, said Lin, who is trial co-chair. Its primary end point is the rate of complete remission without detectable residual leukemia.

Another trial, VINCENT, is testing the combination specifically in patients with NPM1 mutations, one of the groups excluded in the PARADIGM trial. Both myeloMATCH and VINCENT are phase 2 trials.

Lachowiez expects the field to move toward venetoclax-based treatments regardless.

“My leaning,” he said, “is that venetoclax-based therapy probably is going to move into the front line for most AML subgroups, except for those that have very clear phase 3 evidence suggesting [another] regimen that is proven to work.”

PARADIGM was funded by AbbVie and Genentech, which jointly developed venetoclax and provided the drug at no cost for the trial. Fathi reported being a consultant to both companies. Watts reported no relevant disclosures for the cited study. Lachowiez reported having consulting or advisory roles with AbbVie, Genentech, and other companies. Lin reported consulting for Daiichi Sankyo, Jazz Pharmaceuticals, Johnson & Johnson, and other companies.