Nitrosamine impurity risk files for Indian generic makers selling abroad
TL;DR Impuritrace helps generic drug and API makers work out which nitrosamine impurities could form in their products, how potent each is likely to be and what limit applies. Molecular models do the first pass and staff chemists sign off. Regulatory affairs teams pay per molecule or per year.
Ever since nitrosamines turned up in valsartan in 2018, and later in ranitidine and metformin, regulators have treated them as a standing risk across all medicines[1]. The limits are tiny - about 96 ng a day for NDMA and 26.5 ng a day for NDEA[1].
Every product a company sells in the US or EU needs a documented risk review, and newer drug-related nitrosamines each need a potency category[2]. India makes about 20% of the world's generic medicines by volume[3], so its regulatory teams carry thousands of these reviews, often by hand and with outside consultants.
What it does, in regulatory terms: decision support for a company's own regulatory and quality staff. It does not test or release batches, and the drug maker stays responsible for what it files.
A chemist uploads a molecule and its synthesis route. Graph models trained on published structures flag where secondary amines and nitrosating agents could meet, list the likely impurities and score each one against the regulators' structure-based potency method[4]. Where the rules leave room for doubt, the model gives a ranked view and an Impuritrace chemist reviews it. The output is a draft risk file with every step traceable.
I've reframed the seed. A graph-learning platform for discovering new drugs needs years and deep pockets, and it has few Indian buyers. The same modelling skills fit a narrow, paying compliance problem that Indian generic makers face every quarter.
Regulatory affairs and quality heads at mid-size generic and API companies with US or EU filings pay. They pay consultants today, and I couldn't find a public price list. Expect to sell per molecule to smaller firms and an annual licence to firms with large portfolios. Contract research organisations that run impurity work for several clients are a second buyer.
FDA revised its nitrosamine guidance in September 2024 and ties recommended limits to the structure-based potency method[4]. FDA issued separate limits for drug-related nitrosamines in August 2023[2], and an ICH M7 addendum on nitrosamines has been under development since June 2024[1]. Every one of these changes sends teams back to reassess their portfolios.
Nobody publishes a figure for spending on nitrosamine assessment in India. The adjacent number is India's pharma exports, about US$31 billion in FY26 (about ₹2.73 lakh crore)[3]. Treat that as the size of the export book these reviews protect, not as your market. A bottom-up count of filed products per client is the number to build.
Lhasa Limited in the UK sells nitrosamine risk assessment and in silico mutagenicity tools used across the industry[5]. Plenty of firms apply FDA's rule-based potency method by hand, which is free. EU regulators expect each marketing holder to review its own processes[6], so in-house teams and consultants do much of the work. The gap is an India-priced service that pairs models with chemist sign-off and hands over a ready draft file.
| Difficulty | Hard (53/100) |
| Starting capital | ₹1.5–2.8 Cr |
| Time to launch | 12–18 months |
The science is the easy half to describe and the hard half to prove. You need medicinal chemists who can defend every call to an FDA reviewer, validated software with audit trails and enough reference cases to show the model agrees with published categories. Selling into quality teams is slow and careful, and I wouldn't expect that to change.
Impuritrace needs no licence of its own, but its files must meet foreign drug regulators' rules, and the client stays responsible for what it submits.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let the model sign off a file or replace lab testing. A qualified chemist reviews every call, and the drug maker owns what it files.
Want to work out the AI layer for Impuritrace? Talk to Yash →
I'd back a founding pair with a pharma regulatory background and a computational chemist. If you're a pure machine-learning team then it's a poor fit. Prove one thing first - on 50 published drug-related nitrosamines, your categories match the regulators' published ones.
Plan for roughly ₹1.5–2.8 Cr, mostly experienced chemists, model building and the time it takes to win trust from regulatory teams.
No licence of your own was found. Your files must follow FDA and EMA guidance, and a regulatory consultant should confirm Indian requirements.
Yes. The FDA, EMA and ICH rules are global, so any market with generic exporters works. Confidentiality contracts matter everywhere.
The free Impuritrace validation kit has the line-by-line capital breakdown, the effort scoring, the regulatory and software-validation checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for regulatory heads, and 5 assumptions to test. Download the kit →
Thinking about building Impuritrace or something close to it? The make-or-break step is showing a quality head that your files survive regulator questions. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.