Brazil's regulator approved 18 new cancer drugs in five years. The FDA approved 73.

A retrospective comparison published in the Journal of Cancer Policy on 15 May tallied every first-time oncology drug approval issued by the United States Food and Drug Administration, the European Medicines Agency and Brazil's ANVISA between 2020 and 2024, then measured how long each took and what happened afterwards inside Brazil's public health system [s1].

The headline gap is straightforward. The FDA granted 73 approvals over the period, the EMA 56, and ANVISA 18 [s1].

The timing gap

ANVISA's median approval delay was 353 days compared with the FDA and 336 days compared with the EMA [s1]. Convergence between the FDA and EMA was high; ANVISA sat apart from both [s1].

Part of the divergence is evidentiary rather than administrative. The study reports that FDA and EMA decisions relied on a more heterogeneous mix of trial phases and designs, whereas ANVISA more frequently required phase 3 randomised studies [s1]. That is a meaningful distinction. A regulator that more often requires randomised phase 3 data will necessarily decide later on drugs its counterparts approved on earlier-phase or single-arm evidence. Whether that represents a defect or a deliberate evidentiary standard is a judgement the study does not settle, though its recommendations lean toward closing the gap [s1].

What the approvals covered

Across the three agencies, the study found a relatively balanced distribution between biological and synthetic medicines, with approvals concentrated among companies headquartered in the United States and indications concentrated in lung cancer, multiple myeloma and breast cancer [s1]. Nothing in that profile is specific to Brazil; it describes where the global oncology pipeline has been pointed.

The second gap, which is larger

Regulatory approval in Brazil does not mean public availability. Access through the Unified Health System (SUS) requires a further step: incorporation, assessed by CONITEC, the National Commission for Health Technology Incorporation. The study tracked the interval between ANVISA approval and potential SUS incorporation and reports that incorporation remained limited [s1].

This is the structural point the paper is built around: a drug can clear ANVISA and still not reach SUS patients, because registration and incorporation are separate processes handled by separate bodies. The authors' conclusion is that greater integration between regulatory evaluation and health technology assessment is needed to reduce these asymmetries and improve timely access within SUS [s1].

A separate retrospective study published on 28 April puts numbers on that second interval. Analysing biological drugs recommended by CONITEC between 1 January 2012 and 28 March 2024, it found that biologics for rheumatoid arthritis took an average of 2,019 days — roughly five and a half years — from ANVISA approval to CONITEC's incorporation recommendation [s2]. For comparison, the same analysis reports averages of 1,242 days for rheumatoid arthritis and 1,683 days for cancer in England, and 744 days and 1,315 days respectively in Australia, with Australia incorporating faster than both England and Brazil [s2]. The authors conclude that the length of the Brazilian process may hinder access and delay treatment, a barrier they describe as particularly significant for low-income individuals who rely exclusively on SUS [s2].

Stacked end to end, the two studies describe a compounding delay: roughly a year of additional wait at the registration stage relative to the FDA and EMA [s1], then a multi-year assessment interval before a technology is recommended for public financing [s2].

What the study does not establish

The design is retrospective and descriptive, built from publicly available data from the FDA, EMA, ANVISA and CONITEC [s1]. It counts approvals and measures intervals; it does not measure patient outcomes, and it does not establish that faster approval would have produced better survival. That distinction matters in oncology specifically, where a substantial share of drugs approved on surrogate endpoints have not gone on to demonstrate overall survival benefit. A country that approves fewer cancer drugs more slowly is not automatically serving its patients worse — nor automatically better.

The study also does not report the reasons ANVISA did not approve the drugs the FDA and EMA did. A count of 18 against 73 could reflect regulatory conservatism, sponsor decisions not to file in Brazil, agency capacity, or all three; the available analysis does not disentangle them.

The incorporation study carries its own limits. It covers biological drugs only, across two disease areas, and its comparators are the English and Australian assessment bodies rather than the full set of systems Brazil might be measured against [s2]. It reports an average time to a recommendation, which is not the same as the date a patient can collect a prescription. And an average drawn across a twelve-year window can conceal whether the process has been speeding up or slowing down within it — the available reporting does not break the interval down by year.

What to watch

Whether Brazil moves toward parallel or sequential-but-linked review — running health technology assessment alongside or immediately after registration rather than as a fully separate downstream process. That is the specific mechanism the study's policy summary points to as a way to improve timely and equitable access within SUS [s1].