First relapse is the decision that decides everything after it
If you have multiple myeloma and your first-line treatment has stopped working, you are standing on the single most consequential decision point in the disease. Not because first relapse is the last chance — it is not — but because the choice you make here determines which options remain later.
In China, two doors opened at almost the same time. BCMA CAR-T has been available since 2023. In October 2026 the BCMA bispecific teclistamab was approved for use after just one prior line. Both act on the same target, both appear in the same guideline, and only one of them can be used first without weakening the other.
That is the whole article in one sentence.
What changed in October 2026
Teclistamab (泰立珂) has been approved in China since June 2024, but only for patients who had already received at least three prior lines, including a proteasome inhibitor, an immunomodulator and an anti-CD38 antibody. In October 2026 the label was expanded to adults with relapsed or refractory disease after at least one prior line that included an anti-CD38 antibody and lenalidomide — the first-relapse setting.
The expansion is based on MajesTEC-9, and it makes teclistamab the first BCMA-directed therapy in China licensed this early. Every other BCMA product still carries a third-line label.
One detail worth knowing before you compare notes with your oncologist at home: in the United States and Europe the version approved at this treatment line is the combination of teclistamab with daratumumab (MajesTEC-3, approved by the FDA in March 2026 and by the European Commission in August 2026). China's October approval is monotherapy. Different regimen, different data.
The trial behind the label
MajesTEC-9 randomised 593 patients with one to three prior lines, all of whom had been exposed to an anti-CD38 antibody and lenalidomide. Half received teclistamab alone; half received investigator's choice of PVd or Kd.
| Outcome | Teclistamab | PVd or Kd |
|---|---|---|
| Progression-free survival at 18 months | 69.8% | 26.9% |
| Median progression-free survival | not reached | 8.2 months |
| Overall survival at 18 months | 79.2% | 68.6% |
| Response rate | 84.5% | 54.2% |
| Complete response or better | 65.9% | 16.8% |
| Grade 3 or 4 infections | 41.6% | 29.0% |
The hazard ratio for progression or death was 0.29 (95% CI 0.23 to 0.38), and for death 0.60, at a median follow-up of 17.3 months.

Reading those numbers honestly
Five things the headline does not say.
"Not reached" is not "cured." It means fewer than half of the patients on teclistamab had progressed at the cutoff. The number will move as follow-up matures.
The comparison is not teclistamab against nothing. Of the patients in the control arm who went on to receive further treatment, 68.4% received a bispecific or CAR-T. The trial compares immunotherapy now against standard therapy now with immunotherapy later.
It is monotherapy. The daratumumab combination looks better still, and is not what the Chinese label covers.
The infection price is real. Grade 3 or 4 infections occurred in 41.6% versus 29.0% — an absolute increase of more than twelve percentage points, in exchange for the progression-free survival gain.
Nobody in this trial had prior BCMA therapy. The population had one to three prior lines with PI, IMiD and anti-CD38 exposure. If you have already received a BCMA bispecific, this trial's numbers are not your numbers.
For the BCMA CAR-T efficacy numbers themselves — FUMANBA-1, LUMMICAR-1 and CARTITUDE-1, read side by side — see our trial-data analysis.
What the Chinese guideline actually recommends
The Chinese Society of Hematology published a dedicated guideline for first-relapse myeloma in March 2026, and it is unusually concrete. It defines early relapse as progression within 18 months of starting standard three-drug induction, or within 12 months of a frontline autologous transplant. Patients who relapse that early without any high-risk feature at diagnosis are labelled functional high-risk. Early relapse accounts for roughly 10 to 12 percent of all newly diagnosed patients, and the guideline says to treat them as high risk regardless of cytogenetics.
| Situation at first relapse | What the 2026 Chinese guideline recommends |
|---|---|
| Early relapse or functional high-risk | Clinical trial first. If none is available, CAR-T or a bispecific antibody |
| Aggressive relapse (new 17p deletion, 1q21 gain, extramedullary disease, high LDH, short remission) | Same as early relapse: CAR-T and/or bispecific, or a trial |
| Non-aggressive, slow biochemical relapse | Conventional regimen first; CAR-T and bispecific kept for later lines |
| Extramedullary relapse, including plasma cell leukaemia | Trial first, then CAR-T as first choice; dual bispecific may help |
| Frail or over 70 | CAR-T and bispecific remain usable and safe |
There is a gap between a guideline and a label, and you should know where it sits. A recommendation that CAR-T or a bispecific is preferred at first relapse does not mean every hospital will offer it there — some will only do so inside a trial, some off-label with written consent, some not at all. Ask which one you are being offered.
Why the order matters more than the choice
The most useful sentence in the whole guideline is buried in a note about prior treatment: BCMA-targeted therapy given within the previous six months may reduce the efficacy of BCMA CAR-T. The International Myeloma Working Group's sequencing paper says the same thing from the other direction. The guideline also warns that bendamustine or high-dose alkylators within six months can impair T-cell collection and CAR-T expansion, and that prior radiotherapy can compromise stem cell collection.
That produces a real asymmetry, not a preference:
- If you take the bispecific first and it fails, your BCMA CAR-T later starts from a weaker position — and some trials will exclude you outright for prior BCMA therapy.
- If you take CAR-T first, the bispecific is still there afterwards, and a GPRC5D bispecific is a different target entirely.
So in a patient who is a candidate for both, CAR-T first preserves more options than the reverse. That is a sequencing argument, not a superiority argument.

The exception matters just as much. A bispecific can be started in days. CAR-T needs four to six weeks of manufacturing, plus workup and travel. If the disease is aggressive and cannot be held with bridging therapy, waiting for cells is not a neutral choice — the bispecific is the pragmatic one, and the sequencing argument becomes academic.
The three routes, as they actually exist in China
| Route | What it is | Label in China | Typical cost |
|---|---|---|---|
| BCMA CAR-T (eque-cel, zevor-cel, cilta-cel) | One infusion, one manufacturing run | After 3 prior lines | ¥1,150,000 to ¥1,166,000 |
| BCMA bispecific (teclistamab) | Off-the-shelf subcutaneous injection | After 1 prior line since October 2026 | Per vial, self-pay |
| GPRC5D bispecific (talquetamab) | Off-the-shelf injection, different target | After 3 prior lines | Per vial, self-pay |
| Clinical trial | BCMA or GPRC5D CAR-T under development | Eligibility set by the protocol | Drug usually sponsor-covered |
As of 12 November 2026 there were 49 recruiting BCMA CAR-T studies and 19 recruiting GPRC5D studies for multiple myeloma in China on ClinicalTrials.gov. That density is the reason the trial route is a real option here rather than a last resort.
The full product list, prices and target-by-target details are in our guide to the ten approved CAR-T products in China, and the service itself — centres, eligibility review and what we coordinate — is on the CAR-T therapy in China page. What matters for this decision is simpler: the licensed BCMA CAR-T products all sit at third line, while the newest bispecific now sits at first relapse. That mismatch — guideline says go early, label says go late — is the gap a Chinese centre can work inside.

What teclistamab costs, and why the comparison is unfair
Chinese pharmacy retail listings show the 30 mg vial at ¥9,551 and the 153 mg vial at ¥33,246. Dosing is a step-up of 0.06 mg/kg and then 0.3 mg/kg, followed by 1.5 mg/kg weekly, with the interval extended to every two weeks once a patient has been in response for at least six months. For a 60 kg adult, one weekly dose is roughly ¥28,000 to ¥33,000 in drug alone.
That means the first year costs on the order of ¥1.0 to ¥1.2 million, and it continues. A CAR-T course costs ¥1,150,000 to ¥1,166,000 once. One is a course; the other is a subscription. If you are paying out of pocket, that structural difference is the entire economic argument, and it is larger than any single number in a price list.
Two cautions. These are list prices, and reimbursement status for teclistamab in China changes with each national list cycle — check the current position rather than a 2024 price sheet. And in many countries the arithmetic runs the other way: teclistamab is reimbursed and CAR-T is not, which is precisely why patients arrive in China having already used a BCMA bispecific, and why the sequencing question above is not hypothetical.
The wider comparison — including what US insurers do and do not cover — is in our 2026 cost comparison.
The trial route is unusually deep here
Myeloma is the strongest case for the trial route in China, for a boring structural reason: the domestic CAR-T developers and the big haematology centres run a large number of investigator-initiated and industry studies in the same disease, and many are still recruiting in 2026.
What that buys you is a route for the situations where licensed products do not fit: too few prior lines, prior BCMA exposure, extramedullary disease, or a target that has to be GPRC5D because BCMA has already been used. Before you count on it, ask four questions — whether the drug is sponsor-covered or billed, whether screening can be done in English, whether the site accepts foreign patients at all, and which prior therapies exclude you.
What the Chinese team needs from you
Five things shorten this decision from months to weeks.
- Cytogenetics repeated at relapse. FISH for 17p deletion, 1q21 gain or amplification, 1p32 deletion and the t(4;14), t(14;16), t(14;20) and t(11;14) translocations. The guideline is explicit that this must be redone at relapse, because new high-risk clones appear that were absent at diagnosis.
- A whole-body PET-CT. Extramedullary disease changes the recommendation, and it often appears at relapse in patients who never had it.
- A line-by-line treatment table. Drug, dose, dates, best response, why it stopped, and toxicity — not a narrative summary.
- Any prior BCMA exposure, named and dated. Bispecific or CAR-T, including trials. This is the single item that most changes what a Chinese centre can offer you.
- Current organ function and infection screening. Cardiac, renal and liver, plus hepatitis B and C, HIV and syphilis.
The paperwork side — pathology slides, customs approval for human tissue, the visa route — is covered step by step in how CAR-T in China actually works.
Who should not go
This is the section most sites leave out.
- Uncontrolled infection. Not a relative contraindication here. Grade 3 or 4 infections already occur in over 40 percent of treated patients.
- Cardiac function below the centre's threshold. CAR-T and bispecifics both provoke cytokine release; a borderline heart is a real risk, not a checkbox.
- CNS involvement or plasma cell leukaemia. These are urgent, and cell manufacturing takes weeks. If bridging cannot hold the disease, the timeline is the problem, not the choice of product.
- No caregiver for the monitoring period. Two to three weeks of inpatient monitoring after infusion, then weeks of close follow-up, are not optional.
- A plan built on hope rather than on a documented response. If two immunotherapy approaches have already failed and the disease is progressing fast, a third trip is not obviously right.
The honest limits
Myeloma is still incurable. CAR-T produces deep remissions — a complete response or better in roughly three quarters of patients in the pivotal eque-cel study — and most patients eventually progress. The point of choosing the order carefully is to have something left when that happens, not to avoid it.
Second, the guideline's optimism about early relapse exists because the alternatives are poor, not because immunotherapy is guaranteed. Early relapse is high risk by definition; a 69.8 percent progression-free survival at 18 months is a large gain over 26.9 percent and still means a substantial number of patients progress within two years.
Third, trial results come from selected populations. Your age, organ function, prior lines and cytogenetics decide whether those numbers apply to you.
Getting started
If you are at first relapse and considering China, the sequence that wastes the least time is: repeat the FISH and PET-CT locally, put your prior lines and any BCMA exposure into one table, and send those three things for a review before discussing travel or dates. A centre that answers with a product recommendation before seeing your cytogenetics is guessing.
If the decision is to proceed, the operational steps, timeline and payment order are in the patient journey guide. If a first attempt has already failed elsewhere, the options after relapse are laid out in what comes after CAR-T. And if you are comparing this against a transplant, the three doors of myeloma and lymphoma therapy covers that trade-off directly.

