A protected medicine: for a few more years, only one company may sell it.
Photo: Unsplash (royalty-free library image; not the original article's photography)
30 秒速读
先建立框架,再进原文。默认只给英文,点摘要段落(或右上角「译」)出中文对译,点数字卡展开中文。
Merck reports third-quarter results on Thursday with half of its sales coming from one drug, Keytruda. Its main US patent expires in December 2028, and every defence now has a deadline.
The patent cliffis coming forMerck's $31.7 billion drug— and it will not be late
专利悬崖,正在逼近默沙东那支 317 亿美元的药——而且它不会迟到。
PARA 01
Mercktodayreports third-quarter results,and one drug still decides the story, for better and for worse,while the rest of the company is read through it.
默沙东今天公布第三季度业绩,而决定故事走向的仍是一支药,成也它、败也它,其余业务都要透过它来看。
PARA 02
Keytruda, together with its new injectable version,produced$8.4 billion in the second quarter.That wasslightly more than half of Merck's $16.6 billion in sales.
The drug, whose scientific name is pembrolizumab,generated$31.7 billion in 2025, making it the world's biggest cancer medicine.It is the company's largest product by far,and it fundsnearly half of everything Merck does.
A patentisa legal monopoly with an expiry date.In the United States,the main patent on the Keytruda molecule runs outin December 2028, while Europe keeps protection until 2031, and that is the date the industry counts down to.
The trade-off is deliberate:a twenty-year monopolyrewardsa decade of trials that may fail.The clock starts at filing, not at launch, so the years of protection are fewer than they look, and the fall at the end is the patent cliff.1
Chart 1 | The clock starts at filing, not at launch: by the time a drug reaches patients, half the patent term is already gone.Editor's schematic. Term is 20 years from filing; development and approval typically take 10-12 years.
PARA 06
After that date,rivalscan sell copies.For a biologic, the copy is a biosimilar:2similar but not identical, approved on its own trials,and usually sold at a discount, which payers reward because they push patients toward the cheaper version.
Copies do not arrive all at once. First a few makers, then the whole market.仿制品不会一次到来,而是一板一板地来:先是少数厂商,然后是全市场。Photo: Unsplash (royalty-free library image; not the original article's photography)
PARA 07
The industry calls the resultthe patent cliff— loss of exclusivity, in the language of the filings.3Revenue does not slide gently;it dropsas doctors switch prescriptions, quarter after quarter, which is why analysts watch the first down quarter so closely.
Merckprefersa softer metaphor.It describes the transition as “more of a hill than a cliff”,with a shallow dip and a fast return to growth.It is a marketing argument as much as a medical one.
Its first defenceisa subcutaneous version, sold as Qlex.4It is injected under the skin in under two minutes,instead of dripping into a vein for half an hour, which is easier for clinics to schedule.
A treatment room. Moving patients from the infusion chair to a two-minute injection is the core of the defence.治疗室。把患者从输液椅换到两分钟的皮下注射,是默沙东防线的核心动作。Photo: Unsplash (royalty-free library image; not the original article's photography)
PARA 10
Qlexcarriesits own patents on the formulation and the delivery device, and a device patent can outlast the molecule.So if patients switch before 2028,a maker of the old infusioncannot simply move them onto a copy.
The switchis working, slowly:Qlex sold $128 million in the first quarter and $463 million in the second, a line that is rising from a very low base.That is about 5.5% of the franchise.5
Merckwants30% to 40% of US patients on the injectableby the end of 2027.It says penetration is already in double digits, a big shift for a drug usually given in hospitals.
On 7 October,a Dutch courtfoundthat the injectable infringed a patent owned by Halozyme,and blocked itin eight European markets; Germany had already done so in December 2025, and Merck says it is evaluating next steps.
Mercksays it “strongly disagrees”,and arguesthat the patent is invalid.The intravenous version is unaffected,so European patients can still be treated. It is also the version that will face biosimilars first.
The money at stake in Europeis small, and Europe is a smaller market than the United States.The damage is to timing:the company cannot lock in patients therebefore the copies arrive.
The second defenceisa portfolio of new drugs.Winrevair, for a rare lung disease, sold $588 million, up 75%,and Welireg rose 67% to $271 million, both far faster than the company overall.
On paper,the growthlooks impressiveuntil you compare the bases.Each new drug sells less than a tenth as much as Keytruda in a single quarter,and no single product can fill the gap.
Chart 2 | Four products, one quarter: the fastest-growing new drug is still about a fourteenth of the franchise it must replace.Source: Merck second-quarter 2026 results. Figures are reported sales, not annualised.
PARA 18
The third defenceis to buy growth.Merck spent about $16 billion on acquisitions in six months,including $6.8 billion for Terns Pharmaceuticals, partly funded with debt.That is a lot to pay for assets that are not yet on the market.
Those dealsexplaina strange set of numbers.Merck reported a GAAP loss of $0.54 a share,while its adjusted loss, the company's own measure, was only $0.13.
The gapisan acquired research charge of $5.7 billion, or $2.31 a share,booked when Terns was bought.Accountants call it in-process research and development, or IPR&D.6
Such chargesinvolve real cash,but they do not recur like rent.So companies present adjusted earnings,7and investors must judge how adjusted is too adjusted; the direction of travel matters more than any single year.
Competition is not waiting for 2028.In September,India's Ciplasaidthat its US unit would license a Keytruda biosimilar from China's Qilu Pharmaceutical, more than two years before the US patent ends.
Policyadds pressurefrom another side.US Medicare can now negotiate prices on some drugs,which lowers the net price even before a patent expires — a second clock, and a faster one.8
Investorshave noticedboth the risk and the growth, and for now they are pricing the growth.The stock returned 66.9% in the year to 27 October,against 15% for the S&P 500.
Ittradesat 5.3 times sales, against 3.2 for the index.That price-to-sales multiple already assumes the new drugs arrive on time, which is the whole bet.9
Today's reportwill be readfor one number above all:US growth of Keytruda, now that penetration is close to its peak.The patent date, by contrast,cannot be negotiated, while everything else buys time.
to be certain to meet something, usually something unpleasant, at a fixed time
The patent cliff has a date with Merck's best-selling drug.
注定要面对:在某个固定时间一定会遇上某件事,通常是坏消息
专利悬崖和默沙东最畅销的那支药,注定要碰面。
more of a hill than a cliff/mɔːr əv ə hɪl ðæn ə klɪf/
a way of saying that a fall will be gentler than people fear
Executives describe the transition as more of a hill than a cliff.
更像缓坡而不是悬崖:用来安抚说跌幅没有大家担心的那么陡
高管把这次转变描述为更像缓坡而不是悬崖。
lock in/lɒk ɪn/
to fix something in place so that it cannot be taken away later
The company cannot lock in patients before the copies arrive.
锁定:把事情定下来,之后不容易被拿走
公司无法在仿制品到来前锁定患者。
buy time/baɪ taɪm/
to delay something bad so that you have more time to prepare
Everything else buys time; nothing changes the date.
争取时间:把坏事往后拖,好腾出时间准备
其余一切都是在争取时间;没有一个能改变那个日期。
on paper/ɒn ˈpeɪpə/
judged from the numbers alone, before real conditions are taken into account
On paper, the growth looks impressive until you compare the bases.
纸面上:只看数字,还没考虑实际情况
纸面上,这些增长很亮眼,直到你比较基数。
fill the gap/fɪl ðə ɡæp/
to supply what is missing in order to close a difference
No single product can fill the gap.
填补缺口:把差的那一块补上
没有哪个单品能填上这个缺口。
close to its peak/kləʊs tə ɪts piːk/
near the highest level something is likely to reach
Penetration in the United States is now close to its peak.
接近顶峰:接近它能达到的最高水平
美国的渗透率如今已接近顶峰。
a tenth as much as/ə tenθ æz mʌtʃ æz/
one part in ten compared with something else; used to show a very large size gap
Each new drug sells less than a tenth as much as Keytruda.
只有……的十分之一:用来显示两者规模差得很远
每支新药的销售额都不到 Keytruda 的十分之一。
quarter after quarter/ˈkwɔːtə ɑːftə ˈkwɔːtə/
repeatedly, one three-month period following another
Revenue drops as doctors switch prescriptions, quarter after quarter.
一个季度接一个季度:连续不断地
收入随着医生切换处方而下滑,一个季度接一个季度。
读后测验 · Check Your Understanding
3 道题。点选项立即判分;解析需要单独点开。
Q1Which figure best describes Merck's dependence on Keytruda in the second quarter of 2026?
正确答案:B
文章把口径写得很直:Keytruda 与 Qlex 合计 84 亿美元,而公司当季营收 166 亿美元,「略多于一半」。A 把「一半」记成「十分之一」,是把新药与老药的规模搞反了;C 混了季度与年度,也没有「三分之一」这个数;D 文中从未用利润口径描述依赖度。B2 的第一题考的是数字的准确定义,不是大概印象。 Keytruda, together with its new injectable version, produced $8.4 billion in the second quarter. That was slightly more than half of Merck's $16.6 billion in sales.Read the base: $8.4bn against $16.6bn. Change either number and the sentence stops being true.
Q2Why does the subcutaneous version matter so much for the coming patent expiry?
正确答案:D
关键句是「Qlex 带着自己在配方和给药装置上的专利」,所以换过来的患者不能被简单转到旧分子的仿制品上——这是「剂型转换」作为防线的全部逻辑。A 文中没有成本比较,事实上生物药换剂型并不更便宜;B 错在分子没变,仍是帕博利珠单抗;C 与事实相反,新剂型同样需要审批。B2 的第二题考概念辨析:剂型专利不是分子专利,但它能起到争取时间的作用。 Qlex carries its own patents on the formulation and the delivery device. So if patients switch before 2028, a maker of the old infusion cannot simply move them onto a copy.The defence is legal, not chemical: same molecule, different patents.
Q3What does the Dutch ruling most likely cost Merck?
正确答案:C
文章明说「欧洲涉及的钱并不多,真正的损失在时间上」,因为公司无法在仿制品到来前锁定那里的患者。A 夸大:禁令只涉及皮下剂型,静脉版本不受影响,且欧洲只占全球收入的一部分;B 与原文冲突,默沙东明确说静脉版本仍可供患者使用;D 混淆了两件事——美国分子专利 2028 年到期是既定的,与荷兰判决无关。B2 的第三题考推断:读到「钱少、时间贵」,要能推出真正的代价。 The money at stake in Europe is small. The damage is to timing: the company cannot lock in patients there before the copies arrive.The sentence sets up its own contrast: small money, large timing. The answer follows the second half.