Most drugs are a molecule you can copy. A bespoke therapy is not — it is a manufacturing process sold as a product, run once per patient, and the thing you are actually paying for is not a compound but a supply chain that ends inside one person's body. That single structural fact decides who wins the modality, and it decides it in India's favour more cleanly than anywhere else in medicine. Because when the product is the manufacturing, the moat is cost and logistics and process control — not the molecule, not the patent, not the discovery. India cannot out-discover the West. It can out-manufacture it. And in cell and gene therapy it already has: the same class of cure the United States sells for nearly half a million dollars, made in Mumbai for thirty-six thousand, while the American product is still on patent. That is the half of this story that is finished. The other half — gene therapy — India has not built at all, and the reason it hasn't is a wall this essay ends on.
The modality where manufacturing is the product
Start with what makes cell and gene therapy different in kind from every modality that came before it, because the difference is the whole argument.
A small-molecule drug is an object. You discover it once, patent the composition of matter, and then the game is manufacturing a chemical that is identical in every pill, shippable, shelf-stable, and — once the patent lapses — copyable by any competent generics maker. The value lives in the discovery and the patent; the manufacturing is a solved, low-margin afterthought. That is the modality India learned to win last, at the patent cliff, on cost alone.
A bespoke cell therapy inverts this completely. In autologous CAR-T, you take a specific patient's own T-cells, re-engineer them outside the body to recognise their specific cancer, grow them, and infuse them back. There is no vial of product sitting in a warehouse. The "drug" is a process performed on one person's cells, on a clock, under sterility, and delivered back into the same person before the cells lose potency. Gene therapy is the same shape one turn further: a construct built to correct a specific mutation, manufactured and delivered as a bespoke event. In both, the molecule is almost beside the point. What you are selling is the ability to run the manufacturing — reliably, cheaply, at scale, close to the patient.
And that relocates the moat. When manufacturing is the product, the competitive edge is process mastery, vector production, cold-chain and logistics, decentralised operations, and labour cost — every one of which is an execution variable, not a discovery variable. This is the exact axis on which a talent-rich, capital-disciplined, manufacturing-deep country is built to win, and the exact axis on which the West's cost structure is a liability rather than an asset. Bespoke therapy is not one more modality where India can fast-follow. It is the modality whose entire value chain is made of the thing India is structurally best at.
The proof is already on the market.
The half that is won: CAR-T at one-eleventh the price
In October 2023, India's drug regulator approved NexCAR19 — the molecule filed as actalycabtagene autoleucel and now carrying the WHO international name talicabtagene autoleucel; one drug, three names a careful reader should collapse into one — for relapsed or refractory B-cell lymphoma and leukaemia. It was India's, and Asia's, first indigenous CAR-T therapy. It came out of a decade of work at IIT Bombay and Tata Memorial Centre, commercialised through the company ImmunoACT and backed by Laurus Labs.
The number that matters is the price. A CAR-T infusion in the United States lists at roughly $373,000 for Yescarta and $475,000 for Kymriah — and that is the drug alone, before the apheresis, the lymphodepletion, the hospital stay, and the management of the immune storms the therapy can trigger, which together push the real episode toward and past a million dollars. NexCAR19 launched at about ₹42 lakh and has settled near ₹30 lakh — roughly thirty-six thousand dollars, or about fifty thousand fully loaded. Compare like with like — list price against price — and India is delivering the same modality at close to one-eleventh the cost.
That number is not a subsidy or a loss-leader. It is a manufacturing architecture. ImmunoACT makes its own lentiviral vector and plasmids in-house, which removes the single largest cost and IP rent in the entire CAR-T stack — most Western developers rent their vector from a handful of contract manufacturers at ruinous prices. It runs a decentralised, hub-and-spoke, point-of-care model that collapses the cryogenic shipping and cold-chain logistics that inflate the Western episode. Its vein-to-vein time — patient cells out, engineered cells back — runs around twenty days against forty-plus globally. And it does all of this at Indian labour and facility costs. The company reported that it turned profitable in 2025, having treated more than 350 patients across roughly 70 hospitals, and is scaling toward fifteen hundred patients a year. This is manufacturing-as-the-product, executed, banked, and expanding.
This is not the generics play — and that distinction is the whole thesis
It would be easy, and wrong, to file this under India's familiar story: wait for the innovator's patent to expire, then make the thing cheaply. That is not what happened here, and the difference is the point of the entire therapeutics thesis.
CAR-T is not off-patent. Kymriah's patent estate runs to around 2033. Yescarta was the subject of one of the largest patent verdicts in the field — a $752 million willful-infringement judgment in Juno v. Kite. These are innovator-protected, actively litigated products. India did not wait at the cliff. It entered while the modality is still on patent, and it did so not by copying but by building a distinct, independently engineered, humanised construct — same modality, its own intellectual property. That is a stronger position than "we made their molecule cheaper the day their patent died." It is "we built our own version of a proven modality and industrialised it while theirs was still exclusive."
In the language this publication has been building: India did not enter at the patent cliff, where the only edge left is cost and the margin is a commodity's. It entered at the validation bend — the point where the modality's clinical binary has been retired by someone else's dollars (CAR-T works; that question was answered in the West) but the space is not yet commoditised — and it platformed the proven. The variance was paid for abroad. What remained was execution, and execution at Indian cost structure is a different business entirely. The old India play was generics at the end of the curve. This is the same instinct — wait for de-risking, then enter with the cost advantage — moved years upstream, onto a modality still under patent, where the reward is not a commodity margin but a category.
The cost crisis India is walking into
The reason this matters beyond India is that the West has priced cell and gene therapy into a corner. Casgevy, the first CRISPR-based therapy, approved by the FDA in December 2023 for sickle-cell disease, lists at $2.2 million. Zolgensma, the spinal-muscular-atrophy gene therapy, has sat near $2.1 million since 2019. These are, in narrow clinical terms, extraordinary — often one-time, potentially curative. In systemic terms they are unaffordable to almost every health system on earth, including large parts of the American one. A curative therapy that no one can pay for is a scientific triumph and a distribution failure at the same time.
The industry knows this, and its own stated answer to the cost problem is precisely the lever India is already pulling: decentralised, point-of-care manufacturing to strip out the cold-chain and centralised-facility costs. India is not proposing that model. It is running it, at scale, profitably, in a live market. This is the "ceiling is the floor" argument in its most literal form — the thing the West experiences as a ceiling on cell and gene therapy, its own cost structure, is the floor India gets to build on, because India's manufacturing discipline is an asset exactly where the West's is a liability. And it is more than an economic argument; it is a sovereignty one. A country that cannot manufacture its own cell and gene therapies imports its cures at two million dollars a dose or does without them. A country that makes them at thirty-six thousand has bought itself health sovereignty in the most advanced modality in medicine. India, for CAR-T, has.
The half that is not built: gene therapy is a white space, not a victory
Now the honesty the thesis requires, because the temptation is to let the CAR-T win carry a claim it cannot support.
India has not won gene therapy. India has not built gene therapy. There is, as of now, no approved indigenous gene therapy product, no commercial gene therapy company, and no domestic AAV platform. Anyone telling you India is a cell-and-gene-therapy power is telling you half a truth: the cell half is real and shipping; the gene half is a green field with a single fence post in it.
That fence post is worth naming precisely, because it is real and because it proves the essay's own argument. Researchers at CMC Vellore and the Centre for Stem Cell Research, working with Emory University, ran India's first-in-human gene therapy trial for haemophilia A — five patients, bleed-free past fourteen months, published in the New England Journal of Medicine. It is a genuine clinical milestone and a legitimate source of national pride. It is also, precisely, an academic achievement: a hospital-and-institute trial, not a company, not a product, not a platform anyone can buy. And here is the detail that turns it from a footnote into the hinge of the argument — the Vellore team used a lentiviral, ex-vivo approach, engineering the patient's own blood stem cells outside the body. They deliberately did not use AAV, the adeno-associated virus that is the default delivery vehicle for most in-vivo gene therapy. They routed around AAV. The question is why, and the answer is the wall.
The wall is delivery, and it is why the second half is hard
Gene therapy is bespoke manufacturing one turn harder than CAR-T, and the extra turn is delivery. Most in-vivo gene therapy depends on AAV as the vector that carries the corrective gene into the body's cells, and AAV manufacturing is the binding constraint of the entire modality. A single cGMP batch can cost on the order of two million dollars for a couple of hundred litres, translating to roughly ten thousand dollars of vector per dose at high loads. Downstream yields run below fifty per cent — against over ninety for a monoclonal antibody — because separating full, gene-carrying capsids from empty ones is a problem no one has scaled cleanly. The raw materials, the plasmids and the sourcing, are scarce and expensive. This is why the Vellore team reached for lentivirus and the patient's own cells instead: they were engineering around the AAV wall, not through it.
So the structure of India's bespoke-therapy position is now exact. The CAR-T half is won because CAR-T's manufacturing, while hard, is a cell process India has mastered and can run at a fraction of Western cost. The gene-therapy half is unbuilt because its manufacturing runs through AAV, and AAV delivery and production is an unsolved, capital-heavy, yield-poor problem that India has not yet stood up an industrial answer to. The moat that made CAR-T winnable — manufacturing mastery — is exactly the moat that does not yet exist for gene therapy, because the manufacturing itself is not yet solved anywhere at a cost that makes it a volume business. Whoever builds the Indian answer to AAV manufacturing and delivery builds the second half of this essay. That is not a small task tucked into a footnote. It is a company that does not exist yet, and it is the natural next thing for a country that has just proved it can industrialise the hardest modality in medicine — provided someone crosses the delivery wall. That wall is its own thesis, and this essay only names it.
Where it breaks
The disciplined risks, because the essay's own second half is its largest one.
The gene-therapy pillar is aspiration, not achievement. If you take away nothing else, take this: India's bespoke-therapy story is one modality won and one modality merely gestured at. Anyone who reads the CAR-T success as evidence that India is a gene-therapy power has misread the essay. Gene therapy is a white space, and white spaces are hard precisely because they are empty — the AAV manufacturing wall that keeps them empty is not a formality India will clear by wanting to.
The cost ratio has to be defended honestly. The one-eleventh figure holds when you compare list price to price, like with like. It does not hold if you quietly compare India's headline price to America's fully-loaded episode cost, or vice versa — that would be arguing with a rigged ruler. The defensible claim is the clean one: roughly $36,000 against roughly $373,000–$475,000, drug price to drug price.
The cost-down is itself commoditisable. India's CAR-T advantage is manufacturing architecture, and manufacturing architecture can be copied. Decentralised, point-of-care CAR-T is exactly the model the West is now trying to build for its own cost crisis. If India's edge is a process rather than a position, it has to compound into something durable — vector IP, regulatory lead, accumulated process data, scale — or it is arbitrage that others eventually match. One profitable company treating hundreds of patients is a proof, not a moat.
And scale is unproven. Three hundred and fifty patients is a validated beginning, not an industry. The jump from hundreds to the tens of thousands who need these therapies is its own manufacturing and financing valley, and India's reimbursement and health-financing systems for a $36,000 therapy — cheap by global standards, still enormous by Indian household ones — are not yet built. Cheap is not the same as accessible.
Close
A bespoke therapy is a manufacturing process wearing the costume of a drug, and that is why it is the modality where India's real advantage — not discovery, but the industrialisation of a proven thing — bites hardest. India has already proved it, in the most literal way available: it makes a cell therapy the West sells for nearly half a million dollars for about thirty-six thousand, with its own construct, its own vector, its own point-of-care manufacturing, while the Western product is still on patent. That is not the generics play at the end of the curve. It is the platform-the-proven play at the validation bend, executed on the most advanced modality in medicine, and it is finished and profitable.
The gene-therapy half is the mirror image: the same doctrine, the same right to win, and none of it built yet — one academic trial that succeeded specifically by routing around the delivery wall that keeps the field empty. India won the first bespoke modality because its manufacturing was winnable. It will win the second only when someone makes the manufacturing winnable — which means solving delivery. The $36,000 cure is real. The next one is a company that doesn't exist, standing on the other side of a wall made of AAV.
The second modality essay in the Atoms and Cells therapeutics thesis, after "From Payload to Program" (RNA), and the cleanest proof of the "platform the proven" / modality-commoditization doctrine. Receipts verified and dated: ImmunoACT's NexCAR19 (actalycabtagene autoleucel; WHO INN talicabtagene autoleucel) was approved by CDSCO on 13 October 2023 as India's and Asia's first indigenous CAR-T, priced at ~₹30 lakh (~$36,000; ~$50,000 fully loaded) against Kymriah's ~$475,000 launch and Yescarta's ~$373,000 drug-only US list, for a ~1/11 like-for-like ratio; it originates at IIT Bombay and Tata Memorial Centre, is backed by Laurus Labs, makes its lentiviral vector and plasmids in-house, runs a point-of-care model with ~20-day vein-to-vein, and reported profitability in 2025 after 350+ patients across ~70 hospitals. US CAR-T products remain on patent (Kymriah to ~2033; the Juno v. Kite $752M verdict), so this is a same-modality-while-on-patent cost collapse with an independent, humanised construct — not a generics/biosimilar play. The global cost crisis it exploits: Casgevy (exa-cel, first CRISPR therapy, FDA 8 December 2023, $2.2M) and Zolgensma (~$2.1M). India's gene therapy is written as white space by design: no approved product, no commercial company, no AAV platform — only the CMC Vellore / Centre for Stem Cell Research and Emory first-in-human haemophilia A trial (lentiviral, ex-vivo, deliberately not AAV; 5 patients bleed-free >14 months; NEJM 2024/25). The delivery/manufacturing wall on the gene-therapy half is real: AAV cGMP batches ~$2M/200L (~$10k/dose), sub-50% downstream yields, empty-capsid separation, plasmid scarcity — the reason Vellore routed around AAV. Regulatory frame: CDSCO/ICMR-DBT National Guidelines for Gene Therapy (2019); NexCAR19 under the New Drugs & Clinical Trials Rules (2019). All figures are directional and dated; live patient and hospital counts tick upward — reconfirm against ImmunoACT before publishing. The thesis in one line: a bespoke therapy is manufacturing sold as a product, India already won the cell-therapy half at one-eleventh the price while it was still on patent, and the gene-therapy half is a white space one delivery breakthrough away.