ACET Prula-cel: $88M CD20 CAR-T Lupus Primer

Adicet Bio (ACET) closed at $9.43 with a market cap of $88 million. It has one asset that matters, one readout that matters, and both land in September 2026. On September 1 the company completed the last patient visit for its planned Phase 1 analysis of prula-cel in lupus and said topline data will arrive this month. Do not mistake a micro-cap allogeneic CAR-T name for a safe investment. It is a binary bet on 22 lupus patients, and the honest framing is that a clean print re-rates the stock several times over while a safety stumble cuts it in half.

The pivot, stated plainly

Adicet spent its early years on ADI-001, an allogeneic gamma-delta CAR-T, in B-cell lymphomas. The GLEAN Phase 1 study produced a 75% overall response rate and a 69% complete response rate across 16 heavily pretreated patients. Those are respectable numbers, but they were never going to beat the autologous CAR-T leaders or the bispecifics that now crowd relapsed lymphoma. So management repurposed the same cell, renamed it prula-cel, and pointed it at autoimmune disease, where CD20-targeted B-cell depletion has a thinner competitive field and where an off-the-shelf product is a genuine structural advantage. The ticker is still ACET, but the thesis is now entirely autoimmune. The broader setup is covered in the cell therapy investing roundup and the autologous versus allogeneic explainer.

Why CD20 and why gamma-delta

Most autoimmune CAR-T programs chase CD19. Prula-cel targets CD20, which sits further back in the B-cell lineage and depletes the plasmablasts and precursors that a CD19-only therapy can leave behind. The cells are gamma-delta T cells, a subset that does not require harvesting and engineering the patient’s own T cells. That is the allogeneic pitch in one sentence: an off-the-shelf dose with no three-to-four-week manufacturing wait and no need for a hospital bridging stay. In November 2025 Adicet reached FDA alignment to dose lupus nephritis and SLE patients in the outpatient setting, which removes the single biggest logistical reason autologous CAR-T has struggled to scale in autoimmune disease.

The market and why the standard of care is beatable

Lupus nephritis is a roughly $2.6 billion market in 2025, growing near 8.5% a year. The approved options are all chronic, daily or weekly drugs: belimumab (Benlysta), voclosporin (Lupkynis, which booked $216 million in 2024 sales), and obinutuzumab (Gazyva), approved in 2025. None of them achieves drug-free remission. Patients stay on therapy indefinitely, and the calcineurin inhibitor voclosporin carries kidney-function monitoring burden. That is the opening a one-time B-cell depleting therapy is trying to occupy, and it is why the autoimmune CAR-T field exists at all. The bet is not that prula-cel is a better chronic pill. It is that a single deep depletion of CD20-positive B cells resets the disease long enough to be called a remission.

The catch is the field’s central unanswered question: durability. When B cells repopulate, does the disease return? In the rituximab era the answer was usually yes. The CAR-T thesis is that depletion is so deep it produces a more durable reset. Adicet’s 13 patients with at least 12 months of follow-up in the September readout are the first real test of whether an allogeneic gamma-delta CAR-T holds that long.

The data the bull case rests on

In October 2025 Adicet reported preliminary data on seven patients, five with lupus nephritis and two with SLE. All five LN patients achieved a renal response, including three complete renal responses and three DORIS remissions. All seven patients showed rapid, sustained reductions in SLEDAI-2K disease-activity scores and in Physician Global Assessment scores. Safety was clean: two Grade 1 cytokine release syndrome events (28%), zero ICANS, one Grade 1 infection, and no serious adverse events. That is the entire bull case in one sentence: seven for seven on disease-activity improvement with a mild safety profile. The full dataset is in the Adicet October 2025 press release.

The caveat is severe and obvious. Seven patients. Two to nine months of follow-up. That is the smallest sample that can possibly move a stock, and it already did: the shares re-rated on the October data before settling back into the single-digit range where micro-cap CAR-T names trade.

What the September readout needs to show

The planned analysis covers 22 SLE patients with or without LN, all with at least six months of follow-up and 13 with at least 12 months. The trial is registered as NCT06375993. The stock re-rates if three things hold. First, the SLEDAI-2K and renal responses hold out to 12 months, because the field’s core question about any B-cell depletion therapy is whether patients relapse when B cells repopulate. Second, the safety profile stays clean at 22 patients, because one case of grade 3 CRS or a single death would tar the entire allogeneic autoimmune story. Third, the cells show persistence or at least durable efficacy despite being allogeneic, because off-the-shelf cells that clear out in weeks would gut the “durable remission without chronic immunosuppression” pitch.

The safety overhang is real and specific

The entire autoimmune CAR-T sector got a shock when Novartis paused its RAP-cel study after three patient deaths, as covered in the autoimmune CAR-T three-deaths analysis. The read-across is not one-for-one: RAP-cel is autologous and targets CD19, while prula-cel is allogeneic and targets CD20. The mechanisms differ, and so does the dosing context, because prula-cel is now dosed in the outpatient setting with Grade 1 CRS so far. But in an $88 million micro-cap, investors do not price fine mechanistic distinctions. They price fear. If the September data shows any grade 3 or higher event, the shares will not drift down; they will gap.

The competitive map and the valuation anchor

The relevant comparison is not against other allogeneic CD20 programs, because there are almost none. That scarcity is the point of the bet. The cleanest public comp is Cabaletta Bio (CABA), an autologous CD19 CAR-T autoimmune play that trades near a $551 million market cap with zero revenue. That is about 6.3 times Adicet’s $88 million. If prula-cel’s allogeneic CD20 data holds through the September readout, Adicet should close most of that gap on the logic that an off-the-shelf, outpatient-dosed therapy deserves at least as much credit as an autologous one that requires manufacturing and hospital time. For the B-cell depletion mechanics behind that durability question, the CAR-T versus bispecifics piece lays out the reset-versus-maintenance debate.

The re-rating math runs further than Cabaletta’s multiple. A single approval in lupus nephritis would justify a market cap well beyond $551 million, because the addressable population is hundreds of thousands of patients and the standard of care is chronic dosing. That is the asymmetry, and it is why position sizing matters more than the thesis.

The balance sheet

Q2 2026 net loss was $21.4 million against working capital of roughly $103 million (current assets of $120.8 million minus $18.0 million in current liabilities). That is about five quarters of runway at the current burn rate. The September readout arrives well before the cash runs out, but a raise is coming either way. A positive print lets Adicet raise at a much higher price to fund a registrational trial. A negative print forces a raise into weakness. The runway is not the risk. The data is.

The verdict

This is a 1% to 2% position-size trade, not an investment. If the September readout confirms durable 12-month responses with a clean safety profile across 22 patients, the stock re-rates from $88 million toward Cabaletta’s $551 million and beyond, and the allogeneic platform (prula-cel plus the ADI-212 PSMA program) starts to get priced in. If the data degrades or shows a serious safety event, the stock loses half or more. Do not buy the full position ahead of the print. Take a small starter, because the asymmetry is real but so is the 22-patient sample. Every abandoned micro-cap looks like a setup until the safety table comes in.