70年后，在开发流感疫苗这件事上我们为什么进展缓慢？

    今年的流感疫情非常严峻。从去年秋天至今，已经有数千美国人死于流感，其中包括56名儿童。美国疾病控制与预防中心已经发布了流感预警，而现在我们才刚刚进入二月，也就是流感疫情通常会集中爆发的月份。

    流感疫情如此严重的部分原因是，今年的流感疫苗注射效果是近十年来最弱的一次。这或许也可以解释今年的流感疫情为什么如此早地就达到流行病级别。美国疾病控制与预防中心表示，今年的流感疫苗有效率只有23%，远低于往年50%到60%的水平。

    流感疫苗并不是一项高新技术，早在1935年，人类就测试了第一支流感疫苗。仅仅7年后，也就是1942年，美国就开始在美军基地里进行大规模的流感疫苗研究。那么经过了70多年的研究，为什么我们还没有创造出一种持续有效的流感疫苗呢？

    分析流感变种

    每年2月，世界卫生组织都会召集全球流感专家进行研讨，以确定下一季流感疫苗所使用的病毒株。全球科学家和医生都在努力收集和分析当前流行的病毒株，希望找到那些有可能演化为新流感的病毒变种。

    美国疾病控制与预防中心的世卫合作中心负责人杰基•卡茨表示：“这是一个持续不断的过程。一个病毒的变种会衍生出一系列变种。它是一种持续不断的线性进化，但也是很难预测的。想获得一支精确的流感疫苗，关键就在于及时获得病毒样本，这样我们才能及时分析，准备下一年的疫苗。”

    一旦这些病毒株被确定，专家就会针对其中的三到四种，来研制下一年的疫苗。这些病毒株会被大量生产，然后交给葛兰素史克、诺华和赛诺菲等生产厂家来生产和销售——这个过程需要显著的提前期，才能保证疫苗在10月初到达各地的医院。

    去年二月中旬开始的疫苗研制过程，也和往年没有什么差别。来自美国疾病控制与预防中心、美国食品药品监督管理局(FDA)、美国国家卫生研究院和世界卫生组织和其他国际机构的官员在瑞士日内瓦召开会议，确定了下一年度可能在北半球流行的流感病毒株，然后将它们交给生产厂商。（为了给南半球研制流感疫苗，去年9月也召开了另一场类似的峰会。）

    到了去年三月，一种流感病毒的变种突然出现了，科学家们知道它有肆虐的可能，但为时已晚。卡茨表示：“时间是我们研制流感疫苗最大的敌人。”

    疫苗是如何生产的

    据葛兰素史克公司介绍，最普通的流感疫苗生产方法往往会使用鸡蛋，这个过程会持续5个月。几百万只受精的鸡蛋会被用作流感病毒的培养基，然后病毒会被收集、提纯并装进药瓶。在疫苗正式投放全国之前，生产厂家和FDA会测试它们的效能和安全性。

    这个方法从上世纪50年代就开始采用。这是一个缓慢且冗长的过程，尤其是考虑到它面临的挑战。首先，生产过程依赖足够的受精鸡蛋。如果鸡蛋的供给不足，那就会影响免疫血清的产量。其次，有些病毒变种在鸡蛋里成长得并不好，有的还会发生意想不到的变异，从而影响疫苗的效能。

    约翰霍普金斯大学布隆伯格公共卫生学院教授安德鲁•皮克兹表示：“问题的关键在于要改变疫苗的生产方式，它是制约疫苗精确性的一个重要因素。其实，我们确定这些病毒株的速度是非常快的，尤其考虑到它对细节的要求。”

    从2013年1月起，一种新的疫苗生产方式开始进入市场。它的生产速度更快，不需要用鸡蛋作为培养基，而且它的效能可能会更高。这种疫苗被称作“重组蛋白疫苗”，它采用了一种流感病毒蛋白质，这种蛋白质是通过改变一种感染昆虫细胞的病毒的基因而合成的。它可以引发人体的免疫反应，生成保护性抗体。

    FluBlok是第一支也是首支获得FDA认证的重组蛋白疫苗。其制造商Protein Sciences公司的女发言人雷切尔•菲尔伯鲍姆指出：“生产出疫苗只需要几个星期，而不是几个月。另外它还含有三倍的抗原，能起到更好的保护效果。有了这项技术，我们基本上避免了时间上的拖延。”

    重组蛋白疫苗通常需要6到12周的生产时间。在发生流感大爆发的情况下，Protein Sciences只需要3到6个月的时间，就能向全美提供5000万支流感疫苗。而用鸡蛋培养的疫苗要想达到这样的应急产量，则至少需要6个月的时间。

    与时间赛跑

    重组技术通过基因手段在成长速度更快的昆虫细胞中孵化流感蛋白质，从而可以为科学家节省更多宝贵的时间，更精确地确定下一季可能爆发的流感病毒变种。它必定能够帮助科学家更好地研制今年的疫苗。

    皮克兹表示：“我们只需要花四五个星期的时间，就能信心十足地抓住即将流行的流感病毒变种的长尾巴。”

    该技术有可能为我们节省宝贵的时间，但它仍是一项新技术。为了应对今年的流感季节，葛兰素史克公司生产了约2400万支基于鸡蛋培养基生产的流感疫苗，在美国还有其他四家公司也在使用鸡蛋培养基技术。相比之下，Protein Sciences公司今年只推出了30万支FluBlok流感疫苗。

    目前整个行业都在大力投资重组技术，其中也包括葛兰素史克。不过菲尔伯鲍姆表示，大多数厂商在这项技术上都落后了10到15年。Protein Sciences计划明年将FluBlok疫苗的产量提高到120万支，不过对于美国总体的流感疫苗需求量来说，这还只是一个很小的比例。

    科学家的终极目标是研制一种能够治愈所有流感变种的疫苗。《财富》记者埃里卡•弗莱正在深入研究科学家们将如何在接下来的大约10年内实现这个目标。

    直到现在，全球科学家还在认真准备今年二月的流感峰会，他们希望在会上精确地确定明年的流感变种——这可能会拯救几万人的生命。

    皮克兹表示：“在大多数年份里，我们的匹配工作都做得很好，这也是我们能够获得进展的情形之一，但这项工作需要各方齐心协力，不只是科学家，生产和监管环节也必须及时跟上才行。”（财富中文网）

    译者：朴成奎

    审校：任文科

    After 70 years, why aren’t we better at developing flu vaccines?

    This year’s influenza season is a serious one. Thousands of Americans — including 56 children — have died from the flu since last fall. The Centers for Disease Control and Prevention (CDC) has labeled the outbreak an epidemic, and we’re barely into February, the month that typically brings the peak of the season.

    Part of the problem is this year’s flu shot has been one of the least effective in a decade, and it may account for why we are seeing the flu reach epidemic levels so early this season. The current flu vaccine is only 23% effective, compared to between 50% and 60% efficacy for a typical seasonal flu vaccine, according to the CDC.

    The flu vaccine isn’t a newfangled technology. The first human influenza vaccines were tested as early as 1935, and extensive flu vaccine studies were started on U.S. army bases in 1942. So, after more than 70 years of research, why aren’t we better at creating a consistently effective seasonal flu vaccine?

    Analyzing the strains

    The answer to that question begins in February of every year, when global flu experts and the World Health Organization meet as part of an annual consortium to peg the influenza strains for the following season’s flu vaccine

    Scientists and doctors from around the world work diligently to collect and analyze the flu strains currently in circulation, looking to pinpoint mutations that could become ground zero for a new epidemic.

    “It’s a constant process,” said Jackie Katz, the head of the CDC’s WHO collaborating center. “One set of mutations will build on another. It’s a constant sort of linear evolution, but it’s one that is very hard to predict. One of the key features to get an accurate vaccine is having the viruses arrive in time, so we can characterize and analyze these viruses for the coming season.”

    Once those strains are identified, the experts choose three to four to include in the next season’s vaccine. Those strains are then produced and handed off to manufacturers, such as GlaxoSmithKline, Novartis and Sanofi, to produce and distribute — a process that requires significant lead time to ensure the doses reach doctor offices by early October.

    This same process happened in mid-February last year. Officials from the CDC, Food and Drug Administration, National Institute of Health, WHO, and other international bodies, met in Geneva, Switzerland, to pinpoint the strains that would be prevalent in the Northern Hemisphere then handed those off to manufacturers. (A separate but similar summit happens for the Southern Hemisphere in September.)

    The problem came in March when a mutation showed up that scientists knew would wreak havoc, but it was too late. “Time is one of our biggest enemies for the flu vaccine,” said Katz.

    How vaccines are made

    The most common vaccine production method useschicken eggs. This process takes about five months, according to GlaxoSmithKline. Millions of fertilized eggs are used as a culture to grow influenza viruses, which are then harvested, purified and packaged into vials. Manufacturers and the FDA then test for potency and safety before shipping the lot releases around the country.

    This method has been in practice since the 1950s. It’s slow and tedious, especially given the challenges. Firstly, manufacturing relies on the availability of enough fertilized eggs. If egg supplies were ever compromised it would affect our ability to produce enough serum. Second, some flu strains don’t grow that well in chicken eggs and undergo undesirable mutations that affect the potency of a seasonal vaccine.

    “The question is about changing the way these vaccines are made. It’s the real factor limiting the accuracy,” said Andrew Pekosz, a professor at Johns Hopkins University’s Bloomberg School of Public Health. “The speed at which we can identify these strains, especially given the level of detail, is very rapid.”

    As of January 2013, a new vaccine production method hit the market. It’s faster, egg-free and potentially more effective. It’s called a recombinant protein vaccine and uses an influenza virus protein that’s made by genetically altering a virus that infects insect cells. The resulting protein is what triggers the immune response in humans to make protective antibodies.

    “It takes only weeks versus months to produce,” said Rachel Felberbaum, a spokeswoman for Protein Sciences, the maker of FluBlok, the first and only FDA approved recombinant influenza vaccine. “It also has three times the antigens, which helps protect better. We can do that without a time lag because of the technology.”

    The recombinant vaccine typically takes between six to 12 weeks to manufacture, and in the case of a pandemic Protein Sciences is ready to provide 50 million doses to the U.S. government in as quickly as three to six months. Egg-based vaccines would require at least six months for similar emergency output.

    Vying for time

    Recombinant technology — which uses genetic modification to incubate the flu proteins in faster-growing insect cells — could help carve out the vital time that scientists need to more accurately target influenza strains that will circulate during the upcoming season. It certainly would have helped scientists better target this year’s vaccine.

    “All we need is to buy four to five weeks to have a high level confidence of catching the long tail of emerging variants,” said Pekosz.

    The technology has potential to buy that extra time, but it’s still new. For this year’s flu season, GlaxoSmithKline produced about 24 million vaccine doses using the egg-based method, and it is one of five companies producing traditional egg-based flu vaccines in the U.S. In comparison, Protein Sciences released 300,000 FluBlok doses this season.

    The industry as a whole is investing in this recombinant technology, including GlaxoSmithKline GSK -0.09% . However, most manufacturers are still about 10 to 15 years behind, said Felberbaum. Protein Sciences plans to scale up to 1.2 million FluBlok doses for the next flu season, though that’s still a small portion of the overall vaccine quantity needed for the U.S.

    Ultimately, scientists hope for a universal vaccine — one shot that would cure all influenza strains.Fortune’s Erika Fry delved into how scientists are working toward that end goal, which remains nearly 10 years off.

    Until then, scientists around the world are working diligently to be prepared for this February’s influenza summit when they hope to accurately pinpoint next season’s strains — potentially saving tens of thousands of lives.

    “In most years, we have a pretty good match,” said Pekosz. “It’s one of those situations where we could improve, but it takes a really concerted effort. It’s not just science, but manufacturing and regulations.”