Written by a parent, not a doctor. Nothing here is medical advice.

Childhood Cancer Lantern
Search Saved
Protocol / study Active, not recruiting

AALL1731: the study behind your kid's roadmap

Updated September 2026. Citations at bottom of page.

AALL1731 · NCT03914625

Also known as: 1731, COG AALL1731

If your kid has standard-risk B-cell acute lymphoblastic leukemia (B-ALL), you are going to hear “1731” a lot. It is the study that put blinatumomab, or blina, into standard treatment for children. Not every kid on the protocol gets blina, so I have laid out the different 1731 pathways and the numbers that go with each one. I found these figures hard to track down and understand. I hope having them together helps other parents. There are a lot of statistics here, but every child has their own journey. For a parent, the number that matters most is n=1: one child, your kid. These figures cannot tell you how your kid’s treatment will go, but they can help you understand the possibilities.

Before you read on

The 1731 protocol is the document your hospital works from. It belongs to the Children's Oncology Group (COG), the network of hospitals that runs these trials. The latest version is not public, though versions are sometimes attached to journal articles. This page draws on what has been shared about 1731 in journal studies, conference talks and press releases. There is a lot of useful information in those sources, but the protocol changes as COG learns more.1, 2, 18 This page does not replace the protocol your oncologist supplied, and it does not represent COG’s latest version of 1731.

The pathways

Kids are placed in treatment groups after induction, based on how well they are responding. Induction is the first five weeks of treatment and is the same for every kid. During those weeks, the lab collects three results: the leukemia’s genetics, whether leukemia cells were in the spinal fluid, and MRD. MRD means minimal residual disease, a count of the leukemia cells left after treatment. You will see it written as a percentage, such as 0.01%. These results place your kid in SR-Favorable, SR-Average or SR-High. SR means standard risk. The group determines the rest of the two years, including whether blina is part of treatment. Tap the boxes below to explore the pathways. The chart describes children in a study, not what will happen to any one child. “In remission” is the study’s disease-free survival: alive with no relapse and no second cancer at that point in time.1

Interactive pathway · tap any box

Outlook of a representative 100 kids on 1731

These percentages do not all measure the same thing. Some are 1731’s four-year results, some are three-year results, and some, such as an SR-Favorable figure, come from other studies. I have tried to make those differences clear.

100 children · standard-risk B-ALL Induction · the first 5 weeks SR-Favorable 41 of 100 favorable genetics and a fast MRD response SR-Average 35 of 100 neutral genetics, or a slower early response SR-High 23 of 100 MRD still there at day 29, or unfavorable genetics HTS MRD test day-29 marrow · 1 in a million Stronger chemo, then MRD end of consolidation · wk 13 Chemo only no blina 41 children over 95% in remission at 5 yrs earlier study: 98.5% Chemo only HTS found nothing about 4 children 94.9% in remission at 4 yrs (1731, June 2026) Chemo + 2 blina cycles HTS found something about 31 children 95.5% in remission at 4 yrs (chemo alone: 90.4%) Chemo + blina stronger chemo about 22 · 1 leaves 94.1% in remission at 3 yrs (chemo alone: 84.8%)
The exact rules for each group, in one table (in plain words)
GroupGeneticsSpinal fluidDay-8 blood MRDDay-29 marrow MRD
SR-FavorableETV6::RUNX1 or double trisomy 4+10, and no steroids in the month before diagnosisCNS1 or CNS2under 1%under 0.01%
SR-AverageETV6::RUNX1 or double trisomy, with either day-8 MRD of 1% or more or steroids in the month before diagnosis*CNS1 or CNS2anyunder 0.01%
double trisomyCNS1 or CNS2any0.01% to under 0.1%
neutralCNS1anyunder 0.01%
SR-HighETV6::RUNX1CNS1 or CNS2any0.01% or more
double trisomyCNS1 or CNS2any0.1% or more
neutralCNS1any0.01% or more
neutralCNS2anyany
unfavorable (iAMP21, KMT2A, hypodiploidy with fewer than 44 chromosomes, t(17;19))CNS1 or CNS2anyany

I put the NCI PDQ tables 6, 7 and 8 into plain words here. PDQ is the US National Cancer Institute’s summary for doctors. I checked every row against figure 1A of Fries et al, Blood Advances 2026, which reproduces the 1731 chart. The two sources agree on every row. *I worked out one cell from the remaining possibilities. The PDQ’s SR-Favorable table requires no steroids in the month before diagnosis. Its SR-Average table only lists the “day-8 MRD 1% or more” route for favorable genetics. The Fries figure defines SR-Average as “standard risk not otherwise SR-Favorable or SR-High,” so a kid with favorable genetics, fast responses and steroids before diagnosis fits there. That only applies if the child is still eligible for 1731: a steroid course longer than 24 hours in the two weeks before diagnosis, without a blood count taken first, makes a child ineligible.3 Some findings rule out this pathway from the start: leukemia in the testicles, CNS3 (enough leukemia in the spinal fluid to count as disease there) and Philadelphia-chromosome-positive leukemia. Children with Down syndrome have their own version. The table also cannot show every child’s individual journey. Oncologists are empowered to adapt the plan to a child’s medical needs, so a kid can be on 1731 and still have changes specific to them. Where the roadmap and this page differ, the roadmap is the one written for your kid.4, 8, 1

The odds, all together

This is the trial’s main finding. In 1731, SR-Average and SR-High kids were randomized: a computer assigned each child by chance to chemo alone or chemo plus blina, rather than a doctor choosing for them. That gives the treatments a fair comparison by helping balance other differences between the groups. The chart combines all the SR-Average and SR-High kids who were assigned this way. That is how the trial reported its main result, and it is the evidence that adding blina helps.1, 2

In July 2024, the trial’s safety committee saw that kids getting blina were already doing clearly better. This independent group watches the results as they come in. It stopped random assignment two years early because continuing to give half the kids chemo alone would not have been fair. The chemo-only treatment groups, called arms, closed: Arm A for SR-Average and Arm C for SR-High. Kids in those arms who had not reached maintenance were offered one or two blina cycles. Treatment stayed the same for kids already in maintenance. The blina arms, B and D, became standard treatment for SR-High and for SR-Average kids whose HTS test, the more sensitive test for remaining leukemia, found something. That is where the arm letters matter now, though they still appear in the registry and papers. Hospitals adopted blina quickly. In an early 2025 survey of 147 US Children’s Oncology Group (COG) hospitals, 91% used it as standard for SR-Average and 98% for SR-High. Getting blina at home rather than in hospital still depends on where you live and what insurance covers. One large center found that home-care coverage gaps delayed blina by an average of 75 days for affected kids. After the hospital began dispensing the seven-day bags itself, delayed starts fell from 12% of cycles to none.2, 5, 6, 37, 38

Alive and in remission

Four years after the split

ASH 2025 update · data to June 2025 · 1,444 children · median (middle) follow-up 3.5 years

Chemo + blina what kids get now 94.8% Chemo alone the old approach 86.9% 0%25%50%75%100%
7.9 of 100more children alive and in remission at four years with blina (94.8% vs 86.9%)
4.4% vs 12.6%relapsed by four years, blina vs chemo alone
3.1% vs 9.4%relapse involving the bone marrow by four years. This accounts for most of blina’s benefit; spinal-fluid-only relapse was unchanged (1.4% vs 2.2%).

Source: Gupta et al, ASH 2025 abstract 758, Results paragraphs 1 and 2. Disease-free survival ± standard error: 94.8 ± 1.5 vs 86.9 ± 2.2. Hazard ratio 0.41 (95% CI 0.27–0.62).

Who gained the most · four-year update

Blina closed the gap for kids who used to do worse

Leukemia still measurable at day 29 92.8% 82.3% Nothing measurable at day 29 95.9% 89.7% Unfavorable genetics 91.7% 77.1% chemo + blinachemo alonewhiskers: ± standard error · bars start at zero

One finding stood out to me. Kids with leukemia still measurable at day 29 and blina treatment (92.8%) did better than kids with none measured who got chemo alone (89.7%). Unfavorable genetics still mattered, but those kids gained the most of the three groups shown. These are four-year figures for kids alive and in remission, from the ASH 2025 update.2

The blina trade-off

It is important to read the benefits above alongside the risks. Blina’s main downside showed up later than the warnings might lead you to expect. The reactions that sound most frightening happened during the blina cycle itself, and they were rare. The bigger cost came later, as infections during chemo after blina. The two charts separate these risks: reactions during the blina cycle, and infections over the whole treatment.

During the blina cycle itself. Blina directs your kid’s own T cells (a kind of white blood cell) toward B cells, the family of cells the leukemia comes from. When it starts working, the immune system can react all at once. That reaction is called cytokine release syndrome, or CRS. It can cause fever and, in worse cases, low blood pressure or trouble breathing. This is why the first days of the first blina block happen in hospital. Seizures were another concern. The chart counts only severe cases of each.

Severe cytokine release syndrome (grade 3 or higher) 0.3% Seizure 0.7% Bars start at zero · scale 0% to 5%
Share of first blina cycles with each reaction, across all 1731 kids who got blina (ASH 2024 plenary abstract). The scale stops at 5%.

Each bar shows the share of first blina cycles with that reaction. The scale stops at 5% so these small percentages are easier to see. Each reaction occurred in fewer than 1 in 100 first cycles. “Grade 3 or higher” means the reaction needed serious treatment; milder fevers are not counted here. No child died during a blina cycle. But the cycle itself does not tell the whole story.24, 1

After the blina cycle. The second chart covers serious infections over the whole two years of treatment, not just the blina weeks.

SR-Average · chemo + blina 14.8% SR-Average · chemo alone 5.1% SR-High · chemo + blina 20.9% SR-High · chemo alone 17% Bars start at zero · scale 0% to 25%
Share of kids with a serious (grade 3 or higher) bloodstream or central-line infection at any point during the whole treatment, by risk group (NEJM 2025, Table 2). Shorter is better; the scale stops at 25%.

Blue means chemo plus blina; gray means chemo alone. Shorter bars mean fewer infections. Start with the SR-Average pair: adding blina nearly tripled the infection rate. Then look at the SR-High pair. These kids get stronger chemo and had more infections with either treatment; the smaller gap could be due to chance. The extra infections did not occur mainly during blina. They came during chemo afterward. The likely reason is that blina removes healthy B cells along with leukemia B cells, and healthy B cells make antibodies. Fewer antibodies leave a child on chemo less protected against infections. Whether extra protection during those later phases helps is an open question discussed under “On the horizon.”1

The treatment plan, phase by phase

The plan below shows the medicines for each group and the days they are given. The phase names (induction, consolidation, interim maintenance, delayed intensification and maintenance) match your kid’s roadmap. Each phase opens with its goal in plain words. Everyone follows the same basic chemo plan, with two differences. SR-High kids get longer, stronger consolidation (8 weeks instead of 4) and stronger later phases. The blina groups get two 35-day blocks, one before and one after the first interim-maintenance phase. That means everyone except SR-Favorable and SR-Average kids whose high-throughput sequencing (HTS) test found nothing. Everyone finishes two years after interim maintenance I starts. Children with Down syndrome and higher-risk features, called DS-High, have their own version: three blina blocks, a 63-day interim maintenance and a shortened delayed intensification. I was not confident enough in the data to show their plan here.3, 14

The drug regimen, group by group

What the two years look like on each path

Chemo + two blina blocks.

Tap a phase to open its drug calendar. Every medicine links to its own page.

Every asparaginase line reads “pegaspargase or calaspargase pegol”, exactly as the trial’s regimen diagram prints it. Calaspargase pegol has been the standard long-acting asparaginase for US patients under 22 since December 2022. Pegaspargase stopped being available for that age group then. Kids treated before then, and kids in many countries outside the US, get pegaspargase on the same days.18, 19, 20, 26 Phase lengths, drug days and goals come from the trial’s public registry, posted consent forms and regimen diagrams reprinted from the NEJM paper. Each calendar identifies the paragraph or page used. Doses and the adjustments every team makes are deliberately left out.3, 11

On the horizon

None of this is settled. 1731 answered one question, whether adding blina helps, and in answering it opened several more. These are the questions the 1731 researchers themselves have raised, and what has been said in public so far. I have tried to keep three things apart: what has been published, what the researchers have said out loud (in italics), and what I am guessing. Where there is a chart, the paragraph next to it says how to read it.

The next trial has a name: AALL2531.

The only public description so far is one paragraph in a December 2025 review by Dr Mignon Loh, the senior author of the 1731 paper. Here is most of it, with the technical terms explained afterward: “…the introduction of blinatumomab for all patients with childhood ALL may permit a reduction in the more toxic elements of traditional chemotherapy… the COG clinical trial (AALL2531) for NCI SR patients plans to randomize patients to treatments that remove elements of cytotoxic chemotherapy while delivering blinatumomab to all in the hopes of maintaining excellent outcomes… this may also include balancing randomization between IKZF1plus and the PAX5-alt subtype among arms for average and higher-risk patients. Furthermore, patients with ETV6::RUNX1 with the 15q deletion will not be eligible for randomization in the SR-favorable arm, even with a favorable MRD response at the end of therapy.” In plain words, “NCI SR” means standard risk as defined by the National Cancer Institute, the group this page covers. Here, “cytotoxic chemotherapy” is ordinary cell-killing chemo. To “randomize” means assign treatment by chance. IKZF1plus, PAX5-alt and the 15q deletion are genetic sub-types of the leukemia. That is everything that is public. There is no registration on ClinicalTrials.gov as of September 2026, no opening date and no drug list. Plans can change before a trial opens. The picture below is my reading of that paragraph, and it is speculation.

SR-Favorable
SR-Average
SR-High
1731 today
Chemo alone
Chemo + 2 blina blocks
Augmented chemo + 2 blina blocks
AALL2531, as I read the paragraph
A coin toss would assign chemo alone or chemo + blina. ETV6::RUNX1 kids with a 15q deletion would be excluded.
Everyone would get blina; chance would assign full chemo or chemo with parts removed.
Everyone would get blina; chance would assign full chemo or chemo with parts removed.
The top row shows the plan described above. The bottom row is one parent’s reading of one paragraph, not a protocol. A “coin toss” means random assignment. Which chemo parts would be removed is not public.

This raises a question parents often ask: does an SR-Favorable kid benefit from blina? Nobody knows, and it is worth saying that plainly. 1731 never tested it. Dr Rau, one of the trial’s leaders, explained why: that group “does so well with chemotherapy alone that they weren’t eligible for our randomization.” The closest earlier study put them at 98.5% alive and in remission at five years. That leaves little room to show a benefit, and every added drug brings risks. The one trial giving blina to every low-risk child today is in China and has not reported. If AALL2531 opens as described, it would be the first real answer. Researchers are also finding differences within the favorable-genetics group. Findings about the 15q deletion were published in June 2026. About 1 in 22 kids with ETV6::RUNX1 carry it and do less well with treatment. That is why the plan would keep them out of the favorable randomization. No current protocol uses it for sorting.28, 32, 21, 10, 22

Blina instead of chemo, not on top of it.

So far, every result on this page comes from adding blina to the usual chemo. The next idea is to use it instead of some chemo. Two big trials are testing this, and the first has reported results. In Europe, AIEOP-BFM ALL 2017 included 709 high-risk kids and assigned treatment by chance. Half got the usual two intensive chemo blocks after consolidation; half got two cycles of blina instead. The full paper came out in the New England Journal of Medicine in September 2026. The first chart shows the main result.

Two chemo blocks swapped for blina · 358 kids 83% Standard chemo · 351 kids 70.3% Bars start at zero · scale 0% to 100%
Event-free at four years, high-risk kids, AIEOP-BFM ALL 2017 (NEJM, September 2026). Whiskers are 95% confidence intervals. Longer is better.

Each bar shows the share of kids who were event-free at four years: alive with no relapse, no second cancer and no treatment failure. Longer is better. The thin lines, called whiskers, show 95% confidence intervals, a range of plausible values around each result. Here, those ranges do not overlap, supporting a clear difference between the groups. The blina group came out ahead. This is the first time replacing chemo with blina has been shown to work in newly diagnosed kids. But the better result also came with a change in which problems occurred. The second chart shows that trade-off.

Treatment-related infection · blina 23.9% Treatment-related infection · chemo 69.4% Nervous-system side effects · blina 12% Nervous-system side effects · chemo 3.2% Life-threatening events · blina 0.5% Life-threatening events · chemo 4.7% Bars start at zero · scale 0% to 100%
What got better and what got worse in the same trial, during the randomized phase. Share of children with each event; shorter is better.

The second chart works the other way. Each bar shows the share of kids who had a problem, so shorter bars are better. Each pair compares blina in blue with chemo in gray. Infections in the first pair fell to about a third of the chemo group’s rate. Life-threatening events of any kind, in the third pair, fell to about a tenth. But in the middle pair, blina’s neurological side effects (problems with the nervous system, such as tremor, confusion or seizures) rose nearly fourfold. The trade-off was fewer infections and emergencies, but more nervous-system side effects. This trial used a different risk group and a different basic chemo plan from 1731, so its numbers do not carry over to your kid’s plan.

Closer to home, the Children’s Oncology Group (COG) has redesigned its high-risk trial, AALL1732. Both treatment groups now get the same two blina blocks as 1731. The question is whether a second antibody drug can replace two chemo blocks. Its kids are older or have higher white blood cell counts than 1731’s, but get the same stronger, or augmented, chemo plan as SR-High. Both trials remove the second half of delayed intensification.16, 29

How many blina cycles, and the spinal-fluid gap.

In a 2025 review, Dr Rau wrote: “The optimal number and sequence of blinatumomab cycles has yet to be established.” In other words, nobody knows whether two cycles is the right number. 1731 used two because that was the design. The European trial also used two. The Chinese low-risk trial uses two short cycles. Nobody has compared one with two in newly diagnosed kids. Another open question is where relapses still happen, which the next chart shows.

Relapse involving the bone marrow · blina 3.1% Relapse involving the bone marrow · chemo alone 9.4% Relapse in the spinal fluid only · blina 1.4% Relapse in the spinal fluid only · chemo alone 2.2% Bars start at zero · scale 0% to 20%
Relapse by four years, by where it showed up (ASH 2025, all randomized kids). Whiskers ± one standard error. The spinal-fluid pair is not clearly different (p = 0.41). Shorter is better; the scale stops at 20%.

Each pair shows where leukemia came back in 1731 by four years. The top pair covers relapses involving bone marrow, where leukemia lives. The bottom pair covers relapses found only in spinal fluid. Blue is chemo plus blina, gray is chemo alone, and shorter is better. The scale stops at 20%. Blina reduced marrow relapses to about a third, but spinal-fluid-only relapse stayed about the same. That fits with Dr Rau’s explanation that blina “is known to have limited activity in the CNS”. CNS means central nervous system: the brain and spinal cord. This is why chemo into spinal fluid stays in every version of the plan, and why preventing relapse there is the next target. One change is already decided for future Children’s Oncology Group (COG) trials. The extra twice-weekly spinal cytarabine during induction for kids with CNS2 spinal-fluid status did not improve outcomes in earlier trials and will be dropped.31, 2, 34

The HTS question.

This question matters most to SR-Average (standard-risk average) families. MRD means minimal residual disease: leukemia cells left after treatment. Flow cytometry, the standard MRD test, finds about 1 leukemia cell in 10,000. HTS (high-throughput sequencing, sold as clonoSEQ) reads the leukemia’s DNA barcode and finds 1 cell in a million. On 1731, SR-Average kids whose day-29 marrow had nothing found on HTS skipped blina and got chemo alone. This rule came from a 2018 COG study of stored day-29 samples from an older trial. Of 282 standard-risk kids, 56 had nothing found on HTS, and 98.1% of those 56 were event-free at five years on chemo alone, with one relapse. In June 2026, the 1731 team presented its own check of the rule, shown in the first chart.

HTS found nothing · chemo alone · 160 kids 94.9% HTS found something · chemo alone · 352 kids 90.4% HTS found something · chemo + blina · 353 kids 95.5% Bars start at zero · scale 0% to 100%
Alive and in remission four years after consolidation, SR-Average kids, AALL1731 (EHA 2026). Whiskers are ± one standard error. The first two bars are not clearly different (p = 0.37).

Each bar shows the share of SR-Average kids alive and in remission four years after consolidation. The top bar is kids with nothing found on HTS, whom the rule sent to chemo alone. The middle bar is kids with something found on HTS who got chemo alone by random assignment. The bottom bar is kids with something found on HTS who got blina. The whiskers show one standard error, a measure of each estimate’s precision. Two things stand out. First, the top and middle bars are close, with no clear difference (p = 0.37). A p-value describes how often a gap at least this big would appear by chance if the groups truly had the same outcomes. Here, 0.37 means about one time in three. That does not show that the groups are equal. Second, the bottom bar is highest, but those kids were not randomized against the top-bar kids. This is a comparison across groups, not a test of blina in HTS-undetectable kids. The presenters concluded that kids with nothing found on HTS may benefit from blina too.

Dr Gupta offered two reasons the top bar was not higher. The group was small, with few relapses, so there may not have been enough children to show a difference. These kids also mostly had neutral genetics, and a few relapsed late. A clear HTS result may therefore mean less here than in other patients. The second chart shows another finding from the same talk.

Flow found none; HTS above 0.01% · 79 kids 85.4% Flow found none; HTS found below 0.01% · 626 kids 93.9% Bars start at zero · scale 0% to 100%
When the two tests disagree, these are the shares alive and in remission four years after consolidation, with treatments combined. On chemo alone, the pair was 80.0% vs 91.5%; with blina, 89.4% vs 96.4%. Whiskers show ± one standard error. The 705 HTS-detectable kids are included; the 160 undetectable kids are not.

Both bars represent kids whose standard flow cytometry test found no leukemia at day 29. They were called MRD-negative, meaning no remaining leukemia was found by that test. HTS found a trace below flow’s detection level (0.01%) in the bottom group. In the top group, HTS found leukemia above that level that flow had missed. This group did less well with both treatments. The presenters suggested that future studies test stronger treatment for them, as a flow result of 0.01% or more already moves most kids to SR-High (double trisomy uses 0.1%). Dr Gupta said this group made up about 8% of SR-Average kids.

There are two more findings to consider. A blood draw cannot yet replace the day-29 marrow: in 1731’s paired samples, about 1 in 6 SR-Average kids with leukemia found by HTS in marrow had none found in blood. A 2026 study of 49 kids from an older trial who would qualify as SR-Favorable found that three-quarters had a trace on HTS, with no difference in outcome. The authors argue that a low-level HTS result alone is not a reason to give a favorable kid more treatment. A future study would need to test any change to the rule for skipping blina, possibly through random assignment. Nothing has been announced.9, 13, 23, 33, 8

Blina by injection: paused.

A shot under the skin instead of a 28-day pump has been tested in adults, and the first trial in children under 12 opened in December 2025. In August 2026 Amgen, the company that makes blina, reported that both the adult and children’s injection studies “paused enrollment of new patients following a partial clinical hold by the FDA,” and that “discussions are underway with the FDA on a path forward to reopen” them. A partial clinical hold is the FDA (the US drug regulator) telling a company to stop part of a study until a question is answered; what the question is has not been made public. This concerns the experimental injection, not the pump that 1731 kids get.17, 30

Extra protection after blina?

The New England Journal of Medicine (NEJM) authors highlighted infections (see “The blina trade-off” above). They suggested studying extra protection during chemo after blina. Examples include immunoglobulin, an infusion of ready-made antibodies to replace those the removed B cells would have made, or preventive antimicrobials, medicines to prevent infections. The Children’s Oncology Group (COG) nursing recommendations from 2025 say a guideline for immunoglobulin replacement “is underway.” The drug’s US label was updated in October 2025. It added a warning about a rare, severe inflammatory reaction (HLH) when cytokine release syndrome (CRS) is unusual or prolonged, and noted the higher seizure risk in Down syndrome.1, 35, 36

Still to come from 1731 itself: the five-year numbers.

The trial was built around five-year figures. Its registry lists five-year disease-free survival, meaning alive and in remission, as the main result to measure in the randomly assigned groups. It also tracks this in SR-Favorable boys because 1731 shortened their treatment to two years and wants to confirm the figure stays above 93%. So far, we have a three-year full paper and a four-year conference update. We are still waiting for five-year results overall, by risk group and for SR-Favorable (1,635 kids), whose results have never been reported. Results for DS-High (Down syndrome with higher-risk features) were reported in December 2025. Among 93 kids treated on the protocol after induction, 84.8% were alive and in remission at four years. The chart compares their relapse rate with the previous trial’s.

1731 DS-High, three blina blocks · 85 kids 4.7% Previous trial AALL1131 · 235 kids 16.9% Bars start at zero · scale 0% to 20%
Relapse by four years (cumulative incidence), counted from the end of consolidation, in kids with Down syndrome and higher-risk features; whiskers ± one standard error. Shorter is better; the scale stops at 20%.

How to read it: shorter is better and the scale stops at 20%. This is a comparison with an older trial rather than a coin toss, so it is weaker evidence than the randomized charts above, but the gap is large.3, 1, 14

Other important statistics

The rest of the numbers: hazard ratio, confidence intervals, overall survival, subgroup figures, each with where it was read

Each number is the study’s estimate, with its standard error (±) where reported. A standard error shows how precise that estimate is, not what to expect for one child.1 DFS means disease-free survival, the “in remission” measure defined under the pathway. Overall survival, meaning the share of children alive, is included for completeness. The page above emphasizes relapse and remission because relapse is what blina changes.

WhatChemo + blinaChemo aloneWhere it was read
4-year DFS, SR-Average + SR-High94.8 ± 1.5%86.9 ± 2.2%ASH 2025 (2), Results, paragraph 1
Hazard ratio for an event, 4-year update0.41 (95% CI 0.27 to 0.62)ASH 2025 (2), Results, paragraph 1
4-year relapse (cumulative incidence)4.4 ± 0.9%12.6 ± 1.5%ASH 2025 (2), Results, paragraph 1
4-year marrow-involved relapse3.1 ± 0.8%9.4 ± 1.3%ASH 2025 (2), Results, paragraph 1
4-year isolated spinal-fluid relapse1.4 ± 0.5%2.2 ± 0.6%ASH 2025 (2), Results, paragraph 1
4-year DFS, MRD 0.01% or more at day 2992.8 ± 1.9%82.3 ± 2.7%ASH 2025 (2), Results, paragraph 2
4-year DFS, MRD under 0.01% at day 2995.9 ± 1.1%89.7 ± 1.7%ASH 2025 (2), Results, paragraph 2
4-year DFS, end-of-consolidation MRD 0.01 to under 0.1% (SR-High)81.8 ± 11.6% (13)30.8 ± 16.8% (16)ASH 2025 (2), Results, paragraph 3
4-year DFS, CNS2 spinal fluid (blina arms)96.8 ± 2.3%not givenASH 2025 (2), Results, paragraph 3
4-year DFS, unfavorable genetics91.7 ± 3.6%77.1 ± 5.9%ASH 2025 (2), Results, paragraph 4
3-year DFS, SR-Average + SR-High96.0 ± 1.2%87.9 ± 2.1%NEJM 2025 (1), Results "Efficacy", Figure 2
Restricted mean survival time gain (analysis horizon 1,652 days, about 4.5 years)72 days (95% CI 36 to 108)NEJM 2025 (1), Results "Efficacy"
3-year relapse, SR-Average + SR-High3.3 ± 0.8%11.8 ± 1.6%NEJM 2025 (1), Results "Efficacy"
3-year isolated marrow relapse, SR-Average + SR-High1.5 ± 0.5%7.7 ± 1.3%NEJM 2025 (1), Results "Efficacy"
3-year overall survival, SR-Average + SR-High98.4 ± 0.9%97.1 ± 1.1%NEJM 2025 (1), Results "Efficacy"
3-year DFS, SR-Average (835 children)97.5 ± 1.3%90.2 ± 2.3%NEJM 2025 (1), Results "Efficacy by risk group"
3-year relapse, SR-Average2.5 ± 0.9%9.8 ± 2.0%NEJM 2025 (1), same paragraph
3-year overall survival, SR-Average100%98.4 ± 1.0%NEJM 2025 (1), same paragraph
3-year DFS, SR-High (605 children)94.1 ± 2.5%84.8 ± 3.8%NEJM 2025 (1), Results "Efficacy by risk group"
3-year relapse, SR-High4.3 ± 1.4%14.4 ± 2.7%NEJM 2025 (1), same paragraph
3-year overall survival, SR-High96.1 ± 2.0%95.3 ± 2.2%NEJM 2025 (1), same paragraph
Serious bloodstream or line infection over the whole treatment, SR-Average14.8% (52 of 351)5.1% (19 of 376)NEJM 2025 (1), Table 2
Serious bloodstream or line infection over the whole treatment, SR-High20.9% (57 of 273)17.0% (47 of 277)NEJM 2025 (1), Table 2
4-year DFS after consolidation, SR-Average HTS-detectable95.5 ± 1.6% (353)90.4 ± 2.2% (352)EHA 2026 (13), Results
4-year DFS after consolidation, SR-Average HTS-undetectablenot given94.9 ± 2.7% (160)EHA 2026 (13), Results
5-year DFS, AALL0932 low-risk kids (two regimens)98.5 ± 0.9% and 98.8 ± 0.8% · 8 relapses in 603Schore 2023 (10), Results, Figure 2A
5-year event-free survival, HTS-undetectable standard-risk kids, chemo alone98.1 ± 2% (56 kids, 1 event)Wood 2018 (9), Results, Figure 4A

“Restricted mean survival time” tells us how many extra days, on average, children in the blina group spent alive and in remission compared with the chemo-only group. The comparison covered 1,652 days, about four and a half years. The four-year update did not report overall survival. In the NEJM paper, 5 of the randomized children died in remission, all SR-High (2 on chemo alone, 3 on blina). None died during a blina cycle.1, 2, 13

Where I read this

38 sources, numbered to match the small raised numbers, with the section, table or page each figure came from
  1. Gupta S, Rau RE, Kairalla JA, et al. Blinatumomab in Standard-Risk B-Cell Acute Lymphoblastic Leukemia in Children. N Engl J Med 2025;392:875–891. doi:10.1056/NEJMoa2411680 · free author manuscript. Where: Methods "Risk stratification and treatment" (the group rules, arms, blina dose and placement); the Supplementary Material at NEJM.org, which includes a version of the trial protocol; Figure 1 (the enrollment chart the 41 / 35 / 23 shares and the 150 HTS-undetectable kids come from); Results "Efficacy" and Figure 2 (three-year figures); Table 2 (infections); Results "Safety" (deaths in remission).
  2. Gupta S, Rau R, Kairalla J, et al. Blinatumomab mitigates the impact of traditional adverse prognosticators among children with standard risk B-ALL: updated results of COG AALL1731. Blood 2025;146(Suppl 1):758. ASH 2025 abstract. doi:10.1182/blood-2025-758. Where: Methods paragraph (arms closed, crossover offered); Results, paragraph 1 (four-year DFS, hazard ratio, relapse); paragraph 2 (day-29 MRD subgroups); paragraph 3 (end-of-consolidation and CNS2); paragraph 4 (unfavorable genetics).
  3. ClinicalTrials.gov NCT03914625, last updated 30 July 2026. Record. Where: "Arms and interventions" (the seven arm labels); "Detailed description" (every phase length and drug day used in the plan calendars, under the headings INDUCTION, SR / HR CONSOLIDATION, ARM A, ARM B, ARM C, ARM D, DS-HIGH).
  4. National Cancer Institute. Childhood ALL Treatment (PDQ), health-professional version, updated 21 April 2025. Risk groups. Where: section "Prognostic (risk) groups under clinical evaluation," Tables 6 (SR-Favorable), 7 (SR-Average) and 8 (SR-High), plus the CNS1/CNS2 definitions and the "over 95% / 90–95% / below 90%" five-year DFS bands.
  5. Children's Oncology Group. Cause for B-CELLebration: immunotherapy trial closes two years early. 5 Dec 2024. COG news. Where: whole post (the early-closure story).
  6. Zheng DJ, Oranges K, et al. Blinatumomab Care Delivery for Pediatric B-Cell ALL. JAMA Netw Open, 8 Sept 2026. Survey of 147 COG hospitals. Where: Results (91% / 98% standard-of-care figures; home-infusion access).
  7. Smith M, Arthur D, Camitta B, et al. Uniform approach to risk classification and treatment assignment for children with ALL. J Clin Oncol 1996;14:18–24. PubMed. Where: abstract (the age and white-count definition of standard risk).
  8. Fries C, Ji L, Devidas M, et al. Minimal residual disease by high-throughput sequencing in standard risk–favorable pediatric B-lymphoblastic leukemia. Blood Adv 2026;10(17):6027–6031. doi:10.1182/bloodadvances.2026020975 · free full text. Where: Figure 1A (the three-box 1731 risk schema, incl. "SR-Average = not otherwise SR-Favorable or SR-High"); Results (49 SR-Favorable kids, 75% HTS-detectable, no outcome difference).
  9. Wood B, Wu D, Crossley B, et al. Measurable residual disease detection by high-throughput sequencing improves risk stratification for pediatric B-ALL. Blood 2018;131:1350–1359. free full text. Where: Results and Figure 4A (56 of 282 standard-risk kids HTS-undetectable, 98.1% five-year EFS, one event). The 88.2% and 80.1% figures are from the same group's ASH 2016 abstract (Blood 2016;128:1086), results paragraph.
  10. Schore RJ, Angiolillo AL, Kairalla JA, et al. Outstanding outcomes with two low intensity regimens in children with low-risk B-ALL: a report from COG AALL0932. Leukemia 2023;37:1375–1378. doi:10.1038/s41375-023-01870-8 · free author manuscript. Where: Results and Figure 2A (98.5% / 98.8% five-year DFS, 8 relapses in 603); Discussion (the day-8 cutoff difference from 1731).
  11. AALL1731 consent forms (version 5 June 2024), posted on ClinicalTrials.gov: ICF_001, induction, ICF_000, post-induction phases, ICF_002, the blina cycle. Which form is which: ICF_001 is the induction consent for standard-risk and Down syndrome B-ALL; ICF_000 is the consent for localized B-lymphoblastic lymphoma, whose Attachment 1 tables are the standard-risk chemo phases (the same phases as Arm A); ICF_002 is the DS-High consent. The randomized SR-Average and SR-High consents were not posted, so Arm B and Arm D specifics come from the registry text (source 3). Where (PDF page numbers): ICF_001 pages 18–19, induction drug table; ICF_000 page 6, the roadmap diagram, and pages 16–18 (Attachment 1), the consolidation, interim maintenance, delayed intensification and maintenance tables and the "2 years from the start of IM I" rule; ICF_002 pages 8–9, the blina cycle table, bag changes and hospital days.
  12. Indiana University Melvin and Bren Simon Comprehensive Cancer Center. PHO-COG-AALL1731 listing; the University of Iowa listing carries the same arm-by-arm text. Where: the Indiana "Treatment" section and the Iowa "Overview" section, both of which repeat the ClinicalTrials.gov text (used only as a cross-check).
  13. Gupta S, et al. Comparison of end induction flow cytometry versus high-throughput sequencing based MRD in newly diagnosed childhood standard risk B-ALL: a report from COG AALL1731. EHA 2026, abstract S112, 13 June 2026. EHA record. Conference abstract, not yet a full paper. Where: Results paragraph (94.9% / 90.4% / 95.5%, P=0.37, 160 / 352 / 353 kids); Conclusion ("may thus also benefit").
  14. Li AM, Kairalla J, Wang C, et al. Outcomes in children and adolescents with Down syndrome and B-ALL: a report from COG AALL1731. Blood 2025;146(Suppl 1):547. doi:10.1182/blood-2025-547. Where: Methods (the DS-High cohort suspended 29 April 2022 and closed; later DS-High patients taken off protocol therapy); Results (84.8% four-year DFS for the 93 DS-High kids treated on protocol after induction; relapse 4.7% vs 16.9% in the post-consolidation comparison of 85 AALL1731 kids with 235 AALL1131 kids).
  15. Gupta S, McNeer J, O'Brien M, Rau R, Teachey D. The challenge of deintensifying chemotherapy for children and adolescents with B-ALL in the immunotherapy era. Hematology Am Soc Hematol Educ Program 2025;2025(1):229–235. doi:10.1182/hematology.2025000709. Where: the "Selected current efforts" section (the next COG standard-risk trial).
  16. Schrappe M, Locatelli F, Valsecchi MG, et al. Blinatumomab for replacing chemotherapy in pediatric acute lymphoblastic leukemia. N Engl J Med 2026;395(11):1075–1089, 17 Sept 2026. doi:10.1056/NEJMoa2604166 · PubMed. The full paper of the AIEOP-BFM ALL 2017 trial (earlier EHA 2026 abstract S103). Where: abstract, Results (709 randomized; four-year event-free survival 83.0%, 95% CI 77.4–87.4, vs 70.3%, 63.8–75.9; hazard ratio 0.51; treatment-related infection 23.9% vs 69.4%; life-threatening events 0.5% vs 4.7%; neurotoxic events 12.0% vs 3.2%; grade 2 or higher cytokine release syndrome 1.1%).
  17. ClinicalTrials.gov NCT07134088: subcutaneous blinatumomab in children with relapsed or MRD-positive B-ALL (Amgen), started 8 Dec 2025. Record. Also NCT06124157 (AALL2131, the Ph+ trial) and NCT03959085 (AALL1732, high risk). Where: each record's "Study overview" and start date.
  18. Kang J, Kanukollu S, Bevinetto A, Corless R, Redner A. Comparing tolerability and toxicity of calaspargase-pegol and pegaspargase in pediatric leukemia patients. J Pediatr Hematol Oncol, published online 22 Aug 2025. doi:10.1097/MPH.0000000000003098. Where: abstract, first sentence: "As of December 2022, calaspargase-pegol is the standard long-acting asparaginase formulation for patients younger than 22 years of age with acute leukemia."
  19. Matherne M, Panetta JC, Hopp J, Inaba H, Pui CH, Karol SE, Swanson HD. Desensitization to calaspargase pegol appears to be less successful than pegaspargase. Blood Adv 2025;9(7):1492–1496. doi:10.1182/bloodadvances.2024015455 · free full text. Where: Introduction: as of 1 December 2022 pegaspargase became unavailable for patients aged one month to 21.5 years, leaving calaspargase pegol as the only primary asparaginase option for pediatric patients in the United States.
  20. Rau RE. Pediatric-inspired asparaginase regimens for patients with acute lymphocytic leukemia. CancerNetwork, 14 Feb 2024. Article. Where: the paragraphs on pegaspargase ("used in most of our frontline trials over the past several decades", now "only available to individuals 22 years and older" in the US) and calaspargase ("a different linker that results in a much longer duration of activity", "given no more often than once every 3 weeks", "only approved in the United States and for use in individuals 21 years or younger"). Dr Rau is a 1731 investigator.
  21. Rau RE, Gupta S. Video interview: Pediatric B-cell ALL: blinatumomab added to chemotherapy in newly diagnosed disease. The ASCO Post, 2024 ASH Annual Meeting, December 2024. Video and transcript. Where: the transcript. Dr Rau: "Patients with a very low expected risk of relapse were called standard risk favorable, and that group does so well with chemotherapy alone that they weren't eligible for our randomization." Dr Gupta: "What can we start to safely take away? It's a really hard question when you have these kind of excellent outcomes, and it's going to take a lot of consultation within the community."
  22. ClinicalTrials.gov NCT06882057: CCCG-ALL2025 protocol for newly diagnosed low-risk childhood B-ALL (Institute of Hematology and Blood Diseases Hospital, China), started 3 March 2025, recruiting. Record. Where: "Brief summary" (two 14-day blinatumomab courses for every low-risk child regardless of MRD).
  23. Gupta S. Video interview: The COG AALL1731 study. Medimix Hematology, EHA 2026 Stockholm in-depth, 30 June 2026. Video and written summary. Where: the video from 3:13 ("there are two potential reasons that we can think of. One, this is still a relatively small patient group with a low number of relapses so we may have been statistically underpowered. But it may also be that ... in this patient population that is mainly neutral cytogenetics ... the prognostic impact of an undetectable HTS is not quite the same that it would be in other patients"), and the written summary on the same page for the discordant group (about 8% of patients; inferior outcomes; blinatumomab improved them but not to the level of concordant MRD-negative patients).
  24. Gupta S, Rau RE, Kairalla JA, et al. Blinatumomab added to chemotherapy improves disease-free survival in newly diagnosed NCI standard risk pediatric B-ALL: results from the randomized COG study AALL1731. Blood 2024;144(Suppl 1):1. ASH 2024 plenary abstract. doi:10.1182/blood-2024-207450 · ASH program page. Where: Results (per-cycle toxicity: 0.3% of first blina courses with grade 3 or higher cytokine release syndrome, 0.7% with seizures).
  25. Rabin KR, et al. Blinatumomab associated seizure risk in patients with Down syndrome and B-lymphoblastic leukemia: an interim report from COG study AALL1731. Blood 2021;138(Suppl 1), ASH 2021 abstract. Abstract. Where: Results (8 of 47 DS NCI high-risk patients had a seizure during blinatumomab, all over age 10). The closure date (29 April 2022) and "taken off protocol therapy" are from source 14, Methods.
  26. Canada's Drug Agency (CDA-AMC). Blinatumomab (Blincyto) reimbursement review, clinical review report. Canadian Journal of Health Technologies, review PC0365. Article. Where: Clinical Review Figure 3 ("Regimens Used in the AALL1731 Trial in Patients With SR-Average") and Figure 4 ("...SR-High"), each "Reprinted with permission from Massachusetts Medical Society", that is, the NEJM paper's supplementary regimen diagrams. Every asparaginase line reads "PEG-Asparaginase or Calaspargase pegol-mknl" with the days used in the calendars; Figure 4 gives SR-High delayed intensification as 63 days with spinal methotrexate on days 1, 29 and 36, and shows a steroid dose before both blina blocks.
  27. APPHON (Atlantic Provinces Pediatric Hematology Oncology Network). Asparaginase products update, pharmacy presentation, 22 January 2025. Slides (PDF). Where: PDF page 18, the slide on switching from pegaspargase to calaspargase pegol: "Delayed Intensification (DI) cycles extended to 9 weeks (63 days) in high-risk arms to allow for 21 days between calPEG doses." The sister trial AALL1732's registry text (NCT03959085, Arm D interim maintenance 2) gives calaspargase pegol on days 2 and 23 where pegaspargase is days 2 and 22.
  28. Wang'ondu RW, Loh ML. Revisiting novel genomic classifiers in the era of immunotherapy for pediatric B-ALL. Hematology Am Soc Hematol Educ Program 2025;2025(1):252–261. doi:10.1182/hematology.2025000712. Dr Loh is the senior author of the 1731 paper. Where: PDF page 8, the paragraph beginning "Determining whether these more recently described genetic lesions": AALL2531 "plans to randomize patients to treatments that remove elements of cytotoxic chemotherapy while delivering blinatumomab to all"; the IKZF1plus/PAX5-alt balancing; ETV6::RUNX1 kids with a 15q deletion "will not be eligible for randomization in the SR-favorable arm."
  29. ClinicalTrials.gov NCT03959085 (AALL1732), last updated 25 Aug 2026. Record. Where: "Detailed description", primary objective and the Arm D / Arm E paragraphs (both arms receive two 28-day blinatumomab blocks; Arm E replaces Consolidation Part 2 and Delayed Intensification Part 2 with two blocks of inotuzumab ozogamicin).
  30. Amgen. Amgen reports second quarter 2026 financial results, 4 Aug 2026. Press release. Where: the blinatumomab paragraphs of the pipeline update ("A Phase 1b/2 study of subcutaneous blinatumomab in pediatric patients ... has paused enrollment of new patients following a partial clinical hold by the FDA. Discussions are underway with the FDA on a path forward to reopen both subcutaneous blinatumomab studies.").
  31. Hall AG, Rau RE. Blinatumomab use in pediatric B-ALL: where are we now? Blood Adv 2025;9(15):3946–3954. doi:10.1182/bloodadvances.2024014043 · free full text. Where: the "Future directions" and CNS passages ("The optimal number and sequence of blinatumomab cycles has yet to be established"; "Blinatumomab is known to have limited activity in the CNS").
  32. Wang'ondu R, Kairalla JA, Shago M, et al. Chromosome 15q deletions confer inferior outcomes among children with ETV6::RUNX1 B-cell acute lymphoblastic leukemia. Blood Adv 2026;10(12):4109–4112. doi:10.1182/bloodadvances.2026019599 · free full text. Where: the letter's results (1,580 standard-risk ETV6::RUNX1 kids from AALL0932 and AALL1131; 15q deletions in 4.5%, 50 of 1,119 with karyotype, with inferior disease-free survival) and its closing remark that the ETV6::RUNX1 kids randomized to blinatumomab on AALL1731 had a three-year DFS of 100% in both groups, small numbers.
  33. Rau RE, et al. Comparison of immunoglobulin high-throughput sequencing MRD in bone marrow and peripheral blood in pediatric B-ALL: a report from COG AALL1731. J Clin Oncol 2024;42(16 suppl):10014, ASCO 2024 abstract. doi:10.1200/JCO.2024.42.16_suppl.10014. Where: Results (15.9% of SR-Average kids with detectable marrow HTS MRD had undetectable blood HTS MRD; closer agreement in SR-High).
  34. Lee-Miller CA, et al. Adding induction intrathecal cytarabine in newly diagnosed CNS2 B-acute lymphoblastic leukemia does not improve outcomes. Blood Adv 2026;10(12):4184–4192. doi:10.1182/bloodadvances.2026019660 · PubMed. Where: abstract, conclusion ("Future COG B-ALL studies will not include additional intrathecal cytarabine in induction for patients with CNS2 disease").
  35. Kubaney HR, Montgomery KE, Guidry L, et al. Blinatumomab administration in pediatric B-lineage ALL: expert recommendations from the Children's Oncology Group. Clin J Oncol Nurs 2025;29(4):303–312. doi:10.1188/25.CJON.303-312 · free full text. Where: the immunoglobulin section ("Development of guidelines for replacement of IVIG in this population is underway").
  36. Blincyto (blinatumomab) US prescribing information, revised October 2025. DailyMed. Where: sections 5.1 and 5.2 (evaluation for HLH/MAS when cytokine release syndrome is atypical or prolonged; increased seizure risk in Down syndrome and consideration of seizure prophylaxis) and the recent major changes list.
  37. Militano O. Practical considerations for Children's Oncology Group clinical trials. COG Fall Meeting, Betsy Poon pharmacy education series, 17 Sept 2024. Slides (PDF). Where: slides 31–32, describing the 12 July 2024 AALL1731 memo (one or two blinatumomab cycles for chemotherapy-only patients not yet in maintenance; no change for patients already in maintenance).
  38. Oranges K, Duffy KA, McDonough S, et al. Real-world pediatric blinatumomab administration: access to outpatient care delivery and impact of a hospital-dispensed model. Pediatr Blood Cancer 2026;73(7):e70314. doi:10.1002/1545-5017.70314. Where: Results 3.2–3.4 (69 children on frontline blinatumomab; average 75-day delay among those affected by home-care coverage gaps; delayed cycle starts 12% to 0% after hospital-dispensed seven-day bags).

Everything on this page was checked against these sources on 22 September 2026, including a fresh read of the ClinicalTrials.gov record, the National Cancer Institute’s PDQ tables and the three consent PDFs. Conference abstracts (2, 13, 14, 24, 25, 33) and interviews (20, 21, 23) are shorter and less reviewed than full papers; I point out when they provide a figure.