Company history

The merger was defensive. Sankyo’s cholesterol drug Mevalotin had already lost its Japanese patent to generics — domestic sales fell 8% in the year to March 2004 and 19% the year after, ceding the category lead to Pfizer’s Lipitor — and its US patent was due to expire in April 2006. It was still enormously profitable, throwing off roughly ¥130 billion of gross profit, about a third of Sankyo’s consolidated total, so its decline left a hole nothing in the pipeline could fill. At the same time consolidation was reshaping the industry from above: in 2004 Sanofi-Synthélabo took Aventis in a hostile bid to become the world’s number two, and Sankyo — second in Japan but nowhere near the global top ten — sat in plain view as a target.

On 25 February 2005 Sankyo and Daiichi Pharmaceutical, ranked second and sixth in Japan, announced they would combine. Their combined sales of ¥919.1 billion would put them behind only Takeda, ahead of the newly formed Astellas at ¥906.6 billion. Rather than merge outright they chose a two-step structure: first a joint holding company created by share transfer, then a full absorption two years later. Daiichi Sankyo was established and listed on 28 September 2005 with 735 million shares and capital of $453.9M (¥50bn); the Japan Fair Trade Commission cleared the deal after examining overlaps such as non-steroidal anti-inflammatories, where the combined share reached 45%. On 1 April 2007 the holding company absorbed both operating firms. Shoda Takashi of Sankyo became president, Morita Kiyoshi of Daiichi chairman, and the stated aims were ¥150 billion of annual R&D, 2,500 domestic sales representatives, and ¥1 trillion of revenue.

Scale, however, did not follow. Every one of the world’s top ten drugmakers in 2004 sold more than $13 billion of pharmaceuticals; Takeda, Japan’s largest, ranked only fourteenth. Revenue at the merged company was ¥925.9 billion in the year to March 2006 and ¥880.1 billion two years later — flat. What the merger did produce, unintentionally, was a pairing nobody had aimed at: Sankyo’s antibody research and Daiichi’s cytotoxic payload DX-8951 now sat inside the same company. In 2005, the year of the share transfer, four researchers began working on antibody drugs.

The answer to the patent cliff was to buy a second business. A new drug succeeds perhaps once in 20,000 attempts and can cost ¥100 billion; Shoda argued that growth toward 2030 required two eyes — developed and emerging markets, innovative drugs and long-selling off-patent ones. Owning a generics maker would keep the revenue that patent expiry otherwise hands to competitors. In June 2008 Daiichi Sankyo announced the purchase of Ranbaxy Laboratories, India’s largest generics firm, taking about 64% for $4.7B (¥488bn) at ₹737 a share, a 31.4% premium; its footprint widened from 21 countries to 56. The shares rose 6.5% on the news, and Shoda said he would deliberately not manage the acquired company, leaving the founding family’s chief executive in place.

Three months later the US Food and Drug Administration barred imports from two of Ranbaxy’s Indian plants. The United States accounted for 30% of Ranbaxy’s sales; its shares fell 66% to ₹252. In the year to March 2009 Daiichi Sankyo wrote down $3.5B (¥360bn) on the holding and posted a net loss of $2.1B (¥216bn) — its first. A 2011 consent decree and a $500 million settlement with the US Department of Justice did not end it: further plants were barred in 2012, 2013 and January 2014, until all four Indian plants were shut out of the largest generics market in the world. The FDA’s report on the Toansa site listed eight observations, including retesting questionable samples until acceptable results appeared, and flagged two of them in capitals as unchanged since the previous inspection.

In April 2014 Ranbaxy agreed to be absorbed by Sun Pharmaceutical at 0.8 Sun shares apiece, leaving Daiichi Sankyo a roughly 9% minority holder; Nakayama Joji, president since 2010, said he expected to earn the losses back through the partnership. Instead, in April 2015 the company sold the entire Sun stake and left foreign generics altogether. Sun’s share price had nearly doubled since the agreement, delivering a merger gain of about ¥360 billion and some ¥400 billion in cash — which covered most of the roughly ¥450 billion in write-downs, goodwill impairment and settlements before the business ever earned it back. Seven years had gone into a generics market that doubled without them, and Nakayama concluded that fighting on two fronts, in global generics and in innovative drugs, was not winnable.

Retreat left a company with nothing to grow into. Olmesartan, its blood-pressure drug selling about ¥300 billion worldwide, would lose US patent protection in October 2016 and Japanese and European protection in early 2017, and in the US roughly 80–90% of such a market converts to generics within a year. The anticoagulant edoxaban had launched in the US in February 2015 behind three rival products and carried a restriction none of them had. Late in 2015 Nakayama declared a pivot to oncology — a field in which Daiichi Sankyo sold not a single drug, and in which even its own staff doubted him. Buying in was considered and rejected: acquisition prices in cancer had settled into a going rate, and at its size the company could not buy enough attempts to raise its odds.

So it turned inward, to research that had begun in 2005 — the year of the merger itself. Agatsuma Toshinori, then in his thirties, had proposed antibody drugs and been given four people and three years to show something. He deliberately chose antibody-drug conjugates precisely because they were hard and littered with failures: the bond between antibody and payload was unstable, and few drug molecules could be carried per antibody. Colleagues objected on the odds. By 2010 a team of about fifteen, picked across departments, was working on it, and the resulting DXd linker chemistry joined the old Sankyo antibody to the old Daiichi payload — carrying more drug per antibody, stably, than anyone else could.

None of that showed in the numbers. Revenue in the year to March 2018 was ¥960.1 billion against ¥929.5 billion a decade earlier, with operating profit of ¥76.3 billion; a US pain-medicine target was abandoned as the opioid crisis broke, some 280 US sales representatives were cut, and 64% of revenue still came from Japan. The industry openly treated the company as a merger candidate, pairing it with Astellas. Then, on 29 March 2019, Daiichi Sankyo licensed DS-8201 outside Japan to AstraZeneca: $1.35 billion upfront and up to $6.9 billion in total, roughly ¥759 billion, with profits and development and commercialisation costs split evenly and Daiichi Sankyo responsible for manufacture. The deal was struck with trials complete, on the eve of regulatory filing.

Enhertu launched in the US in January 2020 and in Japan that May, and then kept widening. Its original target, HER2-positive breast cancer, covers about a fifth of patients; the HER2-low population the company went on to define covers about half, and the trial reported at the 2022 ASCO meeting in Chicago showed progression-free survival extended by roughly five months and the risk of death halved. On 12 July 2022 the shares hit a year high, and by the end of that month a market capitalisation of about ¥6.9 trillion had carried Daiichi Sankyo past Takeda and Astellas — both larger by revenue — to the top of Japanese pharmaceuticals. The same month a federal court in Texas found the company had infringed Seattle Genetics patents and ordered it to pay about ¥5.7 billion, a claim Manabe disputed. Revenue went from ¥1,044.9 billion in the year to March 2022 to ¥1,601.7 billion two years later, with Enhertu alone contributing about ¥450 billion of it.

A single product is its own risk, so the company sold slices of what followed rather than carrying them alone. In July 2020 it licensed DS-1062 to AstraZeneca for up to about ¥630 billion — struck at Phase 1, far earlier in development than the Enhertu deal. In October 2023 it signed with Merck & Co. on three DXd ADCs, Merck funding 75% of development up to $2 billion per product and half thereafter. Under Okuzawa Hiroyuki, president from 2023 and chief executive from 2025, whole-company acquisitions stopped: in December 2024 the company added Merck’s MK-6070 to the alliance for $320 million and bought the gatipotuzumab intellectual property outright from Germany’s Glycotope for $132.5 million, making it the sixth DXd ADC. Datroway, the second to reach market, was approved in Japan in December 2024, the US in January 2025 and Europe that April, aimed at the hormone-receptor-positive, HER2-negative patients Enhertu does not serve.

The last step was to shed what was merely profitable. Daiichi Sankyo Healthcare — maker of the painkiller Loxonin and the cold remedy Lulu, second in Japan’s over-the-counter market behind Taisho, with about ¥76 billion of revenue, a 17% operating margin and four straight years of profit growth — was sold to Suntory Holdings under a contract signed on 15 April 2026 for ¥246.5 billion, about 3.6 times net assets, transferring in stages between June 2026 and June 2029. It was a business with different customers, a different sales force and different distribution from cancer drugs, and its price bought concentration. Revenue reached ¥1,886.3 billion in the year to March 2025, up 17.8%, and ¥2,123.0 billion the year after; Enhertu alone was running at ¥800 billion, and in June 2025 Okuzawa said the next move was to push it into earlier lines of treatment.