利那卡帕韋 (lenacapavir) 是一種新近獲准的藥物，一次性注射即可預防愛滋病毒長達六個月，其研發需要基礎科學、複雜的化學和毅力
Jon Cohen / 2025年6月20日 / 健康 / 新聞 / 科學
利那卡帕韋透過破壞病毒基因周圍的蛋白質（藍色和紫色）所形成之外殼（稱為衣殼），來預防愛滋病毒感染。 約翰·布里格斯
對許多人來說，它是僅次於愛滋病毒疫苗的最佳藥物。週三，美國食品藥物管理局（FDA）批准了一種預防愛滋病毒感染的新方法：抗反轉錄病毒藥物利那卡帕韋，該藥物只需在腹部注射一次即可提供長達 6 個月的幾乎完全保護。全球衛生官員認為這種藥物可能有助於遏止愛滋病毒新感染的浪潮，目前全球每年仍有超過 100 萬例新感染愛滋病毒病例。
這項批准標誌著吉利德科學公司（專門生產愛滋病藥物的製藥公司）長達二十年曲折的科學研究歷程畫上了句號。利那卡帕韋可以阻斷一種 HIV 蛋白，而科學家最初認為這種蛋白並不是合適的藥物靶點，而尋找一種既強效又穩定的分子花費了很多年的時間，而且所花的資源比吉利德在藥物開發計畫上投入的還要多。吉利德開發團隊的病毒學家 Stephen Yant 表示：「該項目多次幾乎夭折」。
但一位不願透露姓名的開發總監表示，管理層很難終止該項目，因為該團隊距離成功始終「只有一步之遙」。
最終，看似該藥物的「阿基里斯之踵」（致命弱點）——不溶性——卻使其成為一種理想的暴露前預防 (PrEP) 形式。因為它可以在人體內停留數月之久，研究人員表示，它可能對全球疫情的影響比市場上現有的每日服用PrEP 藥丸之療法更大，特別是對於撒哈拉以南非洲的少女和年輕女性而言，由於各種社會和文化原因，她們很難每天服用藥丸。
關於這種藥物的普及範圍仍存在疑問；唐納德·川普總統政府削減愛滋病毒控制力度可能會危及貧窮國家獲得愛滋病毒治療的機會。但承諾是明確的。 「我們有機會真正扭轉新感染的曲線」，PrEP 和 HIV 疫苗倡導組織 AVAC 負責人米切爾沃倫 (Mitchell Warren) 表示。
當利那卡帕韋開始研發時，吉利德已經開發了其他幾種成功的愛滋病藥物。 2001年，該公司推出了替諾福韋 (tenofovir) 藥物，該藥物可以抑制愛滋病毒的一種酶，而這種酶可以將病毒RNA「反轉錄」為DNA，這是感染愛滋病毒的關鍵步驟。 （DNA隨後會將自身拼接到人類宿主的染色體中，並在那裡終生保留）。2003年，吉利德還獲得了針對同一種酶的恩曲他濱 (emtricitabine) 的批准。兩種藥物的組合藥物 Truvada 成為了國際暢銷藥。
但酶抑制劑也有缺點。患者通常每天需要服用幾片藥，而副作用會損害他們的肝臟、骨骼和脂肪代謝。吉利德病毒學家托馬斯·西赫拉日 (Tomáš Cihlář) 回憶道：「我們最初的想法是，讓我們研究一下整個 HIV 複製週期，看看哪裡可能存在一些薄弱點，我們可以通過小分子進行干擾」。
他們選擇的目標是一種叫做衣殼的蛋白質，這並不是一個顯而易見的選擇。酵素具有「活性位點」，即既能結合其他分子又能引發化學反應的區域，這使得它們成為藥物開發商有吸引力的目標。衣殼不是酶，而是一種結構蛋白，它形成保護病毒RNA的外殼。它缺乏一個明確的目標網站。但西赫拉日對猶他大學研究衣殼的生物化學家韋斯利·桑德奎斯特 (Wesley Sundquist) 的研究工作產生了濃厚的興趣，他幫助顛覆了人們對衣殼在病毒生命週期中功能的普遍看法。
研究人員長期以來一直認為衣殼形成一種靜態結構，病毒在突破人體細胞膜後不久就會脫落。 「這完全是錯誤的」，桑德奎斯特說。
在 1999 年的一篇《科學》論文中，桑德奎斯特及其同事顯示，該結構具有令人驚訝的動態性，由五到六個獨立的衣殼蛋白組成，這些蛋白可以自組裝成一個「富勒烯 (註) 錐」。該結構是多孔的，允許將 RNA 轉化為 DNA 所需的分子進入其中。後來的研究顯示，它在進入細胞後很長一段時間內仍能保持完整——它非常靈活，甚至可以擠進細胞核裡——而且它的作用遠不止於保護 RNA。
2003 年，桑德奎斯特的研究小組報告稱，衣殼基因中只需發生幾個突變就能使 HIV 難以感染細胞，這顯示這種蛋白質比預期的更脆弱。 「我們過去認為，好吧，它是一種結構蛋白，你必須用某種東西多次撞擊它，才能真正扭曲它的結構，才能產生效果，」他說。但這項研究明確指出，「衣殼蛋白做了很多事情，而且它是這樣，只要你稍微打擾它們，病毒就會在意。 「也許衣殼終究是「可用於藥物治療的」。
一年後又出現了另一個令人鼓舞的跡象。研究人員了解到為什麼愛滋病毒不能感染猴子：它們製造了一種蛋白質，阻止錐體「脫殼」並釋放病毒 RNA。范德比爾特大學的病毒學家克里斯托弗·艾肯 (Christopher Aiken) 一直在研究衣殼在感染中的作用，他說：「那時，每個人都對衣殼產生了興趣」。2006 年 5 月，吉利德允許 Cihlář 組建團隊尋找針對衣殼的藥物。
最初，他們專注於破壞衣殼蛋白組裝成錐體的過程，這是新病毒從受感染細胞中出現的關鍵步驟。他們篩選了 100 萬個小分子來研究它們是否能夠破壞這個過程，並發現了大約 1000 個有希望的候選分子。但所有這些藥物都存在效力、穩定性或被人體分解的方式問題。
隨後，在 2010 年 1 月的一次 HIV 會議上，Cihlář 看到了 Aiken 和輝瑞公司的科學家發布的一張海報，上面描述了一種新發現的衣殼抑制劑 PF74，該抑制劑在 HIV 生命週期的另一端發揮作用。輝瑞團隊並未對其效力所感動，但 Cihlář 渴望盡一切努力維持其項目，因此看到了新的線索。
Aiken 研究小組已經證明 PF74 可以透過多種方式干擾 HIV 的早期生命週期。它不僅導致衣殼過早脫殼，還會干擾反轉錄並與控制感染的衣殼結合細胞蛋白質相互作用。吉利德團隊製作了晶體結構，使其能夠看到分子如何與衣殼蛋白結合，這為研究人員提供了改進 PF74 的路線圖。 「我們製作了大約 50 種類似物，並且能夠將效力提高約 50 倍，」Yant 回憶道。他們也使它更加穩定。
在接下來的 6 年裡，吉利德團隊又創造了 4000 種化合物，最後確定了一種名為 GS-6207 的化合物，這種化合物非常有效且穩定。「我們確實已經將其推向了極限」，Cihlář 說道。
他說，到那時，位於加州福斯特城的公司總部的季度管理評估已經變得困難。 「我認為我們對成功沒有太大的信心」。這些疑慮反映出另一個看似難以克服的障礙：由於 GS-6207 不易溶於水，用它製成的口服藥丸會穿過腸道，而不會被吸收進入血液。 「這就是化學家所說的『磚灰』」，揚特說。
但在一次審查會議上，有人表示糟糕的溶解性可能會使其具有持久力。為了驗證這個概念，研究小組將藥物溶解在聚乙二醇和水的混合物中，然後注射到老鼠的皮下。注射劑在皮下形成一個儲存器，緩慢釋放藥物，使血液中的藥物濃度保持在高水平至少 12 週。 「這些數據讓我們震驚不已」，Yant 說道，並再次拯救了這個計畫。吉利德的化學家進一步對該藥物進行了改進，使其溶解度進一步降低。
2018 年，該公司開始對健康志願者和愛滋病毒感染者進行臨床試驗，以找到合適的劑量並確保這種後來被命名為利那卡帕韋的藥物的安全性。
利那卡帕韋作為一種治療方法很有前景：它的效力至少是目前最好的抗反轉錄病毒藥物的 10 倍（以完全抑制病毒複製所需的劑量來判斷）。但抗愛滋病毒藥物必須合併使用，以防止抗藥性病毒，而利那卡帕韋沒有明顯的合作夥伴，因為沒有其他愛滋病毒藥物可以在一次服用後持續 6 個月發揮作用。因此，吉利德專注於利基市場，將利那卡帕韋注射劑作為其他抗愛滋病藥物治療失敗患者的「挽救治療」。 2022 年 12 月，FDA 批准了該藥物用於此目的。
但挽救性治療的效益並不高，因此吉利德公司押注 lenacapavir 也能成為有效的預防藥物。兩項針對不同族群的大型預防試驗檢驗了這一前景——由於利那卡帕韋的顯著功效，去年這兩項試驗均被提前停止：一項研究顯示，對感染的保護率為 96%，另一項研究顯示，對感染的保護率為 100%。 「結果非常驚人」，協助進行這兩項試驗的開普敦大學傳染病專家琳達·蓋爾·貝克 (Linda Gail-Bekker) 說。 「我被震撼了」。這些研究是本週 FDA 批准的基礎。
桑德奎斯特表示，許多其他公司可能會停止開發世界上第一種衣殼藥物的長期努力。他說：「吉利德公司的人值得高度讚揚，他們堅持解決問題，並用驚人的科學手段解決了問題」。
現在的問題是，最需要幫助的人是否能夠從這種堅持中受益。
譯註：富勒烯（英語：Fullerene） 或巴克球、巴基球（英語：Buckyball）是一種完全由碳組成的中空分子，形狀呈球型、橢球型、柱型或管狀。 它的結構與石墨相似，但除了六元環外，還包含五元環，有時還包含七元環。 根據碳原子數的不同，富勒烯有不同的種類，例如C60、C70等。
註：Lenacapavir PrEP 被《科學》雜誌評為 2024 年之年度突破。鑑於 Sundquist 對開發的貢獻，他於 5 月 8 日與吉利德的臨床開發、HIV 預防和兒科的副總裁 Moupali Das，以及非洲愛滋病毒和愛滋病預防倡導組織的聯合創始人兼執行董事伊薇特·拉斐爾 (Yvette Raphael)， 一起被授予美國科學促進會 Mani L. Bhaumik 年度突破獎。這裡有一個關於他們貢獻的故事，獲獎者也參加了這個小組討論。 《科學》新聞部沒有參與 Bhaumik 獎得主的評選。
doi: 10.1126/science.zqscgug
關於作者
喬恩·科恩
喬恩‧科恩 (Jon Cohen)，《科學》雜誌資深記者，在加州大學聖地牙哥分校獲得科學寫作文學士學位。您可以透過 Signal 的 bval31.65 和 Bluesky 的 @cohenjon.bsky.social 聯繫他。
Always ‘one atom away’: The long, rocky journey to an HIV prevention breakthrough
Developing lenacapavir, the drug newly approved to protect against HIV for six months in one shot, took basic science, sophisticated chemistry, and perseverance
Jon Cohen / 20 Jun 2025 / Health / News / Science
Lenacapavir prevents HIV infections by disrupting the proteins (blue and purple) that form a shell, known as a capsid, around the genes of the virus. John Briggs
To many, it is the next best thing to an HIV vaccine. On Wednesday, the U.S. Food and Drug Administration (FDA) approved a new way to prevent HIV infection: the antiretroviral drug lenacapavir, which provides almost complete protection for 6 months with a single injection in the abdomen. Global health officials think the drug might help quell the tide of new HIV infections, still more than 1 million around the world each year.
The approval caps a tortuous, 2-decade-long scientific journey for Gilead Sciences, a pharma company specializing in HIV drugs. Lenacapavir blocks an HIV protein that scientists initially thought was not a suitable drug target, and finding a molecule that was both potent and stable enough took many years and more resources than Gilead had ever invested in a drug development program. “The project almost died multiple times,” says Stephen Yant, a virologist on Gilead’s development team.
But management had difficulty killing the project, says one director of development who did not want to be named, because the team was always “one atom away” from success.
In the end, what looked like the drug’s Achilles’ heel—its insolubility—made it an ideal form of pre-exposure prophylaxis (PrEP). Because it lingers in the body for so many months, researchers say it may have more impact on the global epidemic than the daily PrEP pill regimens already on the market, especially for teenage girls and young women in sub-Saharan Africa who, for a variety of social and cultural reasons, have had difficulty taking daily pills.
Questions remain about how widely the drug will be available; cuts by President Donald Trump’s administration to HIV control efforts could imperil poor countries’ access. But the promise is clear. “We have an opportunity to truly bend the curve of new infections,” says Mitchell Warren, who heads AVAC, an advocacy group for PrEP and HIV vaccines.
Gilead had developed several other successful HIV drugs when lenacapavir’s development began. In 2001 it introduced the drug tenofovir, which cripples an HIV enzyme that “reverse transcribes” viral RNA into DNA, a vital step in infection. (The DNA then splices itself into the chromosomes of its human hosts, where it remains for life.) In 2003, Gilead also won approval for emtricitabine, which targets the same enzyme. Truvada, a combination of both drugs, became an international blockbuster.
But the enzyme inhibitors had downsides. Patients often had to take several pills a day, and side effects were harming their livers, bones, and fat metabolism. “The idea started, well, let’s look at the whole HIV replication cycle and see where there could be some weak points where we could interfere with a small molecule,” remembers Gilead virologist Tomáš Cihlář.
The target they chose, a protein called capsid, was not an obvious choice. Enzymes have “active sites,” regions that both bind other molecules and trigger chemical reactions, which makes them attractive targets for drug developers. Capsid is not an enzyme but a structural protein, which forms a jacket protecting the viral RNA. It lacks a clear site to target. But Cihlář became intrigued by the work of Wesley Sundquist, a biochemist at the University of Utah who studied capsid—and who helped upend common wisdom about its function in the viral life cycle.
Researchers had long thought capsid forms a static structure that came off shortly after the virus breached the membrane of a human cell. “This is all wrong,” Sundquist says.
In a 1999 Science paper, Sundquist and colleagues showed that the structure is surprisingly dynamic, made up of groups of five or six individual capsid proteins that self-assemble into a “fullerene cone.” The structure is porous, allowing molecules needed to convert RNA into DNA to slip inside. Later work showed it also stays intact long after entering the cell—it’s so flexible it even squishes into the cell’s nucleus—and does much more than protect RNA.
In 2003, Sundquist’s team reported that just a few mutations in the capsid gene could make it difficult for HIV to infect a cell, suggesting the protein was more vulnerable than expected. “We used to think, OK, it’s a structural protein, you’d have to hit it a lot of times with something that really distorts the structure to have an effect,” he says. But the study made clear that “the capsid protein does a lot of things, and it does them in a way that, if you perturb them a little bit, the virus cares about it.” Maybe capsid was “druggable” after all.
A year later came another encouraging sign. Researchers learned why HIV can’t infect monkeys: They make a protein that prevents the cone from “uncoating” to release the viral RNA. At that point, “Everyone became interested in the capsid,” says Christopher Aiken, a virologist at Vanderbilt University who explored its role in infection. In May 2006, Gilead allowed Cihlář to form a team to hunt for a drug targeting capsid.
Initially, they focused on disrupting the assembly of capsid proteins into a cone, a step crucial for new viruses to emerge from an infected cell. They screened 1 million small molecules for their ability to disrupt the process and found about 1000 promising candidates. But all had problems with potency, stability, or the way they would likely be broken down by the body.
Then, at an HIV conference in January 2010, Cihlář saw a poster from Aiken and scientists at Pfizer that described a newly discovered capsid inhibitor dubbed PF74, which worked at the other end of the HIV life cycle. The Pfizer team was unimpressed by its potency, but Cihlář, eager for anything to keep his program alive, saw a new lead.
Aiken’s group had shown that PF74 interfered with HIV’s early life cycle in several ways. Not only did it lead the capsid to prematurely uncoat, it also interfered with reverse transcription and interacted with a capsid-binding cellular protein that controls infection. The Gilead team made crystal structures that allowed it to see how the molecule bound to the capsid proteins, which gave the researchers a road map to improve PF74. “We made about 50 analogs, and we were able to make about a 50-fold improvement in potency,” Yant recalls. They also made it more stable.
Over the next 6 years, the Gilead team created an additional 4000 compounds before settling on one named GS-6207 that was both extremely potent and stable. “We really pushed it pretty much to the limit,” Cihlář says.
By then, quarterly management reviews at the company’s headquarters in Foster City, California, had become rocky, he says. “I don’t think there was a lot of confidence we were going to succeed.” The doubts reflected another seemingly insurmountable obstacle: Because GS-6207 didn’t easily dissolve in water, an oral pill made from it would pass through the intestinal tract without being absorbed and entering the bloodstream. “It was what the chemists called ‘brick dust,’” Yant says.
But at one of the review meetings, someone suggested the lousy solubility could give it staying power. To test the concept, the team dissolved the drug into a mixture of polyethylene glycol and water and injected it under the skin of rats. The injection formed a reservoir under the skin that slowly released the drug, leaving high levels in the blood for at least 12 weeks. “The data just floored us,” Yant says—and rescued the project, again. Gilead chemists tweaked the drug further to make it even less soluble.
In 2018, the company started clinical trials in both healthy volunteers and people living with HIV to find the appropriate dose and make sure the drug, later named lenacapavir, was safe.
Lenacapavir was promising as a treatment: It is at least 10 times more potent—judged by the dose needed to fully suppress viral replication—as the best antiretrovirals in use. But antiretrovirals must be used in combination to prevent the emergence of resistant viruses, and lenacapavir had no obvious partners because no other HIV drug works for 6 months with one dose. So Gilead focused on the niche market of using lenacapavir injections as a “salvage treatment” for people who were failing on other anti-HIV drugs. In December 2022, FDA approved the drug for that purpose.
But there’s not much money in a salvage treatment, so Gilead bet on lenacapavir turning out to be a powerful preventive, as well. Two large prevention trials in diverse populations tested that promise—and last year both were stopped early because of lenacapavir’s remarkable power: One study showed 96% protection against infection and the other 100%. “The results were spectacular,” says Linda Gail-Bekker, an infectious disease specialist at the University of Cape Town who helped conduct both of the trials. “I was blown away.” Those studies are the basis for this week’s FDA approval.
Sundquist says many other companies would have shut down the long-flailing effort to develop the world’s first capsid drug. “The Gilead people deserve enormous credit for sticking to a problem and solving it with amazing science,” he says.
The question now is whether the people who need it most can benefit from that perseverance.
Note: Lenacapavir PrEP was Science’s 2024 Breakthrough of the Year. For his contribution to its development, Sundquist was awarded the AAAS Mani L. Bhaumik Breakthrough of the Year Award on 8 May, along with Moupali Das, vice president, clinical development, HIV prevention and pediatrics at Gilead; and Yvette Raphael, co-founder and executive director of Advocacy for Prevention of HIV and AIDS in Africa. A story about their contributions is here, and the laureates also participated in this panel discussion. Science’s News department had no role in selecting the Bhaumik award winners.
doi: 10.1126/science.zqscgug
About the author
Jon Cohen, senior correspondent with Science, earned his B.A. in science writing from the University of California, San Diego. He can be reached on Signal at bval31.65 and on Bluesky at @cohenjon.bsky.social.
