Pharma Patent Litigation in Indian High Courts 2021-2026 (Dataset)
How pharma patent suits move through the Indian High Courts, what a fair dataset must count, and which three figures matter: suits by molecule, court and outcome, the share of interim injunctions granted, and the median days from filing to first order. Last verified October 2026.
Dataset · Pharma Patent Litigation
Anyone who has to advise on a drug patent fight in India asks the same three questions: how many suits are there, how often does the court grant an interim injunction, and how long does the first order take. Most answers online are anecdotes built from a handful of famous cases. This page explains how pharma patent suits actually move through the High Courts, how a fair dataset should be counted, and which three figures are worth quoting. Last verified October 2026.
- Three figures matter: suits by molecule, court and outcome; the share of interim injunctions granted; and the median days from filing to first order.
- Count rules come first: one suit, one row; molecule recorded as pleaded; pending is not a loss.
- Always read N: a percentage without a sample size and export date is not evidence.
- Last verified: October 2026.
01Why pharma patent data is hard to trust
Pharma patent litigation in India is talked about a lot and counted very little. The reason is simple: there is no single public table that lists every drug patent suit, its molecule, its court and its result.
Famous cases skew the picture
Landmark rulings, such as the Supreme Court decision in Novartis AG v. Union of India (2013) on Section 3(d) of the Patents Act, 1970, get written about for years. Hundreds of routine interim orders do not. A reader who learns the law only from landmark cases will overestimate how often a court writes a long reasoned judgment and underestimate how many suits end quietly in a consent order or a withdrawal.
One molecule, many suits
A single drug often draws several suits: one against each generic maker, sometimes in more than one court. If you count suits and not molecules, one blockbuster can make a year look busier than it was. A useful dataset counts both.
Outcomes are not tidy
Many suits have no final judgment. They end with an interim order, a settlement, or a transfer. Calling every such suit a win or a loss for the patentee is wrong. The outcome label has to match what the record actually shows.
A pharma patent statistic is only as good as its counting rule: what counts as one suit, what counts as one molecule, and what counts as an outcome.
02How a pharma patent suit moves through a High Court
A pharma patent suit in India usually starts as a commercial suit for infringement, often with an application for an interim injunction. Knowing the steps tells you which dates to measure.
The suit and the forum
A patentee sues for infringement of rights under the Patents Act, 1970. Suits are filed in a High Court with original jurisdiction or a district court, as the Act and the rules of that court allow. In the Delhi High Court, patent matters sit in the Intellectual Property Rights Division (IPD), which works under the Delhi High Court Intellectual Property Rights Division Rules, 2022 and a separate set of Patent Suit Rules. Patent matters in that court carry IPD-PAT case-type codes.
The interim injunction
The patentee usually asks for a temporary injunction under Order XXXIX Rules 1 and 2 of the Code of Civil Procedure, 1908. The defendant answers with a challenge to validity, often built on prior art or on Section 3(d), which limits patents on new forms of known substances unless they show improved efficacy.
Other routes that shape the data
Some pharma disputes never reach a suit. A challenger may file a revocation or opposition at the patent office, or seek a compulsory licence under Section 84. The 2012 order in Bayer v. Natco on sorafenib is the best known example. These are not infringement suits, so a clean dataset keeps them in a separate bucket and says so.
For the injunction stage in the Delhi court, see our guide to interim injunctions in Delhi patent suits. For appeals that arise after the IPAB was wound up, see Section 3(k) appeals after the IPAB.
03What a fair dataset must count
A fair pharma patent dataset states its rules before it states its numbers. These are the rules this dataset follows.
- Window: suits filed or first heard between 1 January 2021 and the export date, which is stated on the page.
- Courts: the High Courts that hear patent suits, with the Delhi High Court IPD counted by its case-type codes and other courts counted by case title and subject.
- One suit, one row: a suit that moves between courts or gets renumbered is counted once.
- Molecule: the active ingredient named in the pleadings or order, recorded as written, then grouped by hand where spelling differs.
- Outcome: one of a small fixed set: interim injunction granted, interim injunction refused, settled or withdrawn, final judgment, pending. A suit with no order yet is pending, not a loss.
- Days to first order: counted from the filing date to the first substantive order on interim relief, not to the first listing.
Why we state N every time
A percentage with no sample size is not evidence. Every figure on this page must be read with its N and its export date. Where a figure is not yet frozen, we say so instead of guessing.
04The three figures we report
The dataset is built to answer three questions, each tied to a dated Claw export of High Court records.
| Figure | What it tells you | How it is counted | Status |
|---|---|---|---|
| N suits by molecule, court and outcome | How big the field is, where it is fought, and how suits end | One row per suit, grouped by molecule, court and outcome label | Reported with N and export date once frozen |
| Share of interim injunctions granted | How often the patentee gets interim relief when it asks | Granted divided by all decided interim applications, excluding pending | Reported with N and export date once frozen |
| Median days from filing to first order | How long a patentee waits before the court says anything of substance | Median, not average, because a few very slow suits distort an average | Reported with N and export date once frozen |
We deliberately use the median for time. An average can be pulled up by a handful of suits that sat for years, and it then describes almost no real case.
Until those figures are inserted with their export date, treat any pharma injunction percentage you see elsewhere, including from us, as unverified.
05How to read the numbers
Once you have the figures, four habits keep you from misreading them.
- Look at N first. A grant rate from a few dozen decided applications can move a lot with one more order.
- Split by court. Courts differ in volume, bench practice and speed. A pooled number hides that.
- Split by molecule type. A suit over a new chemical entity differs from one over a new form, salt or combination, which is where Section 3(d) arguments cluster.
- Do not predict one case from the average. Strength of the patent, quality of prior art and delay by the patentee decide an individual injunction application.
If you are planning strategy around these numbers, our guide to AI litigation strategy tools in India covers how teams use case data in practice.
06Where Claw fits
Claw is an all-in-one legaltech platform for Indian advocates, law firms, and corporate legal teams, combining AI-based case search, an AI legal assistant (Legal GPT), case management, and compliance automation across all Indian courts and tribunals.
A dataset like this is only possible if you can pull every suit by court, molecule and outcome from a reliable record. Claw tracks cases, orders and hearings across 9,102 courts, benches and forums, with 30 crore+ judgments and results under 5 seconds, so you can search pharma patent suits by court, party or advocate name and read the interim orders that most counts miss. Its judgments run from 1950 for the Supreme Court and from 1990 for the High Courts, and Claw does not use customer documents to train AI models.
Create a FREE Claw account to search pharma patent litigation by court or party name. To search the wider body of law, see case search across Indian courts.
07Sources and further reading
Primary sources behind this explainer. Check statute text and rule numbers against them before relying on this page.
- Delhi High Court IPD Rules, 2022: Indian Kanoon text and the Delhi High Court website
- Patents Act, 1970 and patent office records: IP India
- Claw: clawlaw.in
Last verified October 2026. Competitor names are mentioned only to state where they lead. No competitor pricing is quoted on this page.
08Frequently asked questions
How many pharma patent suits are there in Indian High Courts?
The count depends on the window and counting rules, so it must be quoted with N and an export date. This dataset covers 2021 to 2026 and counts one row per suit, grouped by molecule, court and outcome. The frozen figure is published with the dataset once confirmed.
How often do courts grant interim injunctions in pharma patent cases?
There is no honest single number without a stated sample. The share is calculated as granted divided by decided interim applications, with pending ones excluded, and it should always be read with N and the export date.
Why use the median days and not the average?
A few suits that sit for years can pull an average far above what most patentees face. The median shows the middle case, which is a fairer guide to what to expect.
Which court hears most pharma patent suits?
The Delhi High Court IPD is the best known forum for patent suits, but the split across courts is a result the dataset measures. See the court breakdown once the export is frozen.
Are compulsory licence and revocation matters counted as suits?
No. They are patent office or appeal matters, so a clean dataset keeps them in a separate bucket from infringement suits.
Can Claw find patent suits by court and party name?
Yes. Claw tracks litigation across Indian courts with a case website, including the Delhi High Court IPD by case-type code and by court, number and party name. It is court-side only for copyright.