The landscape around high-risk smoldering multiple myeloma has been shifting quickly in the past year. Last November, the FDA approved the monoclonal antibody daratumumab (Darzalex Faspro) based on the phase 3 AQUILA trial, which showed the drug delayed progression compared with active monitoring. Since then, early-phase trials of CAR T-cell therapy and bispecific antibodies have reported deep remissions in the smoldering setting — with some investigators raising the possibility of a cure.

But having something to offer patients has made the question of who to offer it to more consequential than ever. At the 2026 American Society of Clinical Oncology Annual Meeting, organizers converted a planned debate on smoldering myeloma into an education session, saying the topic had become too complex for a for-and-against format.

The central question — whether and when to treat — comes down to two problems, said Mateo Mejia Saldarriaga, MD, a hematologist/oncologist at Weill Cornell Medicine and NewYork-Presbyterian in New York City.

"Can we identify the patients who are going to progress?” he said. “And does treating them now actually change the course of disease?"

The Identification Problem

Smoldering myeloma is a precursor condition in which abnormal plasma cells are present in the bone marrow, but they are not causing organ damage. Biologically, it sits in between monoclonal gammopathy of undetermined significance (MGUS) and overt myeloma, said Irene Ghobrial, MD, of Dana-Farber Cancer Institute and Harvard Medical School in Boston.

That in-between space is broad. Some patients with smoldering myeloma progress to active disease within a few years, while others are monitored for 15 or 20 years without progression — making risk stratification essential.

The first step, however, is to ensure the patient doesn’t have early myeloma. That requires baseline imaging with whole-body diffusion-weighted MRI — not the low-dose CT or skeletal surveys that many patients still receive, said Rajshekhar Chakraborty, MD, of Columbia University Irving Medical Center in New York City, and Ghulam Rehman Mohyuddin, MBBS, of the Huntsman Cancer Institute, University of Utah, in Salt Lake City.

Without it, patients with undetected bone lesions may be classified as smoldering when they actually have active disease.

Once early myeloma is ruled out, the most widely used tool for risk stratification, the 20/2/20, classifies patients as high, intermediate, or low risk for progression based on three measures of disease burden: whether abnormal plasma cells exceed 20% of the bone marrow, abnormal protein in the blood exceeds 2 g/dL, or the free light chain ratio exceeds 20. “High-risk” is typically defined as the presence of at least two of those factors.

But even among high-risk patients, only about half progress to active myeloma within several years.

"If you tell me that only 50% of patients will progress in the next 4 or 5 years, that means there's still a significant number of patients who will not progress that you may spare from treatment," Mejia Saldarriaga said.

The risk models also depend on a diagnostic boundary that some experts consider unreliable. Mohyuddin called the line between smoldering and active myeloma a “subjective boundary.” One of the thresholds — a serum free light chain ratio of 100 — can place patients on opposite sides of the diagnosis from one blood draw to the next.

"Sometimes they're above 100, then the next time you check labs, they're below 100, and this continues," Mohyuddin said.

In a perspective published earlier this year, he and Chakraborty argued that the light chain ratio threshold, which was adopted in 2014, substantially overestimates the likelihood of overt myeloma. Independent validation studies, including a Danish population-based cohort, have shown the 2-year risk of progression was as low as 30%.

The instability of the diagnostic boundary is one problem. A separate limitation is that risk models like 20/2/20 rely on a single snapshot rather than tracking where a patient's disease is heading.

“A patient may be high risk at 20/2/20, but then they've been flat for 5 or 6 years, and you start wondering, ‘Are they really high risk?’” Ghobrial, of Dana-Farber, said. “And vice versa, they may be intermediate risk, but their numbers are going up every single time you see them.”

To help address that issue, she and her colleagues developed a dynamic risk calculator called PANGEA, reported on earlier this year in Nature Medicine.

The tool works with standard blood tests — M-protein, free light chain ratio, creatinine, and hemoglobin — without requiring genomic profiling, advanced imaging, or even a recent bone marrow biopsy. And it tracks whether lab values are trending in the wrong direction over serial visits, rather than capturing a single timepoint. 

What Daratumumab Data Showed — and Didn't

Along with the challenges in identifying patients at high short-term risk of progression, the ultimate impact of treating those individuals remains to be seen.

Chakraborty said he applies four questions to any potential treatment for smoldering myeloma: Does it improve overall survival? Does it prevent serious complications like fractures or kidney failure? Does it cure or eliminate the precancerous clone? And is it safe enough for asymptomatic patients?

Applied to daratumumab, the answers look discouraging on the first three counts. The overall survival data from AQUILA were not yet mature, Chakraborty said, and the comparison was compromised because many patients in the monitoring arm did not receive daratumumab-based therapy when they later progressed.

"We really don't have any credible data showing that treatment with single-agent daratumumab improves overall survival," Chakraborty said, noting that a delay in progression is distinct from preventing progression.

In addition, he said, "You're not really preventing them from going onto dialysis or having fractures.” 

The drug was relatively well tolerated in the trial, but not without cost. About 30% of patients had serious adverse events, with grade 3 or 4 infections seen in 16%, driven in part by the drug's immunosuppressive effects. The treatment also requires regular injections for up to 3 years — a significant commitment for someone who is otherwise feeling well.

"This approval [of daratumumab] may do more harm than good," Chakraborty said, as community oncologists may simply prescribe it because a patient meets the label criteria without the nuanced discussion the data warrant. 

Ghobrial, however, said that imperfect risk-stratification is not a reason to withhold therapy.

"We always either over- or undertreat some of our patients, and we will never be perfect,” she said. “But that doesn't mean that we stop treating our patients."

The Case for Treating Early

According to Ghobrial, earlier treatment makes sense because the immune environment of smoldering myeloma offers a therapeutic advantage that disappears once the disease progresses.

She compared the situation to breast cancer. "You never tell a woman, ‘Wait until you have metastasis everywhere, and then I’ll treat you.’”

Some of the strongest evidence for treatment comes from Ghobrial's own research program, called PRISM, which tests different immunotherapy approaches in smoldering myeloma. 

In one arm, the CAR-PRISM trial, a single infusion of the CAR T therapy ciltacabtagene autoleucel produced undetectable residual disease in all 20 high-risk patients in phase 2, with almost no serious toxicity. Although side effects such as immune reactions and drops in blood counts were common, they were milder than typically seen when the same drug is used in advanced myeloma.

In the ImmunoPRISM trial, patients with high-risk disease were randomized to either standard myeloma treatment (lenalidomide combination therapy) or the bispecific antibody teclistamab. Estimated 2-year disease-free survival was 92% among teclistamab patients vs 51% among those in the comparison group — with fewer serious side effects than the same drug produces in advanced disease.

"Patients were flying in from all over the world for this trial," Ghobrial said, referring to CAR-PRISM. "It wasn't us trying to get them on the trial. This was the other way around."

She described patients who were not willing to wait. "I have seen patients tell me, 'Are you waiting for the fractures to happen? Can I be treated now? I have 50% cancer cells in my bone marrow. What are you waiting for?'”

The Bottom Line

Despite the divisions, all four experts converged on one point: Finite-duration bispecific antibody therapy is the most promising path forward. Chakraborty, who still practices active surveillance for most of his patients, said the bispecific data is shifting his thinking.

"If I can do a fixed duration of treatment with minimal toxicity and I can cure maybe 80% or 90% of patients, that will be good enough for me," he said.

Mohyuddin independently agreed: "I could get behind a finite-duration, highly potent drug that is probably curing people, especially for people who have smoldering myeloma that is worsening."

For clinicians navigating this now, Mejia Saldarriaga said that “day zero” decisions are not necessary: Get the best imaging available, track the trajectory of the patient's labs, and have an honest conversation about what the data show and what they do not.

Ghobrial reported financial relationships with Amgen, Bristol-Myers Squibb, Janssen, Sanofi, and others. She is a founder of and holds equity in Predicta Biosciences and is a founder of 4 Degree Therapeutics. Chakraborty reported relationships with Alexion (AstraZeneca), Janssen, Sanofi, and others. Mejia Saldarriaga reported relationships with Johnson & Johnson, Legend Biotech, Sanofi, and others. Mohyuddin reported no relevant financial relationships.