primary endpoint: covid infection rate between placebo and vaccine groups
所有的trial, pfizer, AZ, Moderna的 primary endpoint 都 不是比较住院率，疾病严重程度。而是有任何轻微症状 （咳嗽，头痛，发烧，腹泻，e t c) 的受试者中核酸检测阳性的人数和比例。所以真正完全没有任何症状的人不进入统计也是非常合理的设计，因为有无症状是疫苗是否有效的判断。因为疫苗不会让人不接触病毒 （有可能你的鼻粘膜上有病毒的核算片段，不是活病毒），而是打了疫苗以后可以令病毒无法在你接触以后保持活性和扩增，进而染病。
To confirm influenza A and/or B disease, （条件1）a positive reverse transcriptase polymerase chain reaction (RT-PCR) result for influenza A or B virus from a nose and throat swab obtained concurrently with an influenza like illness (ILI) （条件2）was required. ILI was defined as the presence of an oral or axillary temperature ≥ 37.8 degrees Celsius (°C) in the presence of at least one of the following symptoms on the same day: cough, sore throat, runny nose or nasal congestion.
19% or 95%? US expert challenges Pfizer vaccine’s efficacy, trig
19% or 95%? US expert challenges Pfizer vaccine’s efficacy, triggers debates in China By Leng Shumei and Bai YunyiPublished: Jan 13, 2021 10:55 PM
An illustration shows vials of a COVID-19 vaccine and syringes with the logos of US pharmaceutical company Pfizer and German partner BioNTech. Photo: AFP
Questions raised by Peter Doshi, an assistant professor of pharmaceutical health services research at the University of Maryland School of Pharmacy, have triggered heated discussion on Chinese social media.
Doshi's opinion was disputed among Chinese vaccine experts interviewed by the Global Times. Many experts agreed that Pfizer should provide more raw data for peer review but said Doshi's methodology may be flawed.
Chinese experts also noted that disputes often arise from clinical trial data of COVID-19 vaccines, given the difference in participant groups and how different pharmaceutical companies define COVID-19 symptoms differently. They said that the efficacy rate of a vaccine should be evaluated in a more comprehensive manner, citing Sinovac as an example.
Sinovac vaccine received a 50.3-percent efficacy rate on Wednesday, based on phase III clinical trials in Brazil. But the figure was largely due to the fact that participants in Brazil are all medical workers with a very high infection risk, experts said.
Doshi on January 4 released an article on a blog platform under the UK pharmaceutical journal The BMJ, questioning Pfizer's efficacy rate.
Pfizer revealed that it discovered 170 PCR confirmed COVID-19 cases during the phase III clinical trials and 3,410 suspected cases in total.
However, if taking these suspected cases all as confirmed ones, the Pfizer vaccine's efficacy would be dramatically reduced to 19 percent. Even after omitting cases occurring within seven days of vaccination, which should include the majority of symptoms due to short-term vaccine reactogenicity, the efficacy rate remains as low as 29 percent, Doshi said in the article.
Doshi also questioned the standards of how Pfizer excluded cases and the influence of the use of medication on the vaccine's efficacy.
Doshi's comments triggered controversy on Chinese social media platforms, with some Chinese experts doubting Doshi's counting method for efficacy as he takes all participants showing symptoms to be confirmed COVID-19 cases, even if PRC tests show negative results.
"The PRC test is of very high sensitivity, 98.6 percent in the last year. But its sensitivity could be only 5 percent under Doshi's counting method," Zhuang Shilihe, a Guangzhou-based vaccine doctor, told the Global Times on Wednesday.
Zhuang noted that Doshi's doubts may be related to the anti-vaccine movement in the US, noting that Doshi has previously criticized other vaccines like the flu vaccines.
A Beijing-based vaccine expert who required anonymity expressed distrust of Pfizer's efficacy, but he did not directly comment on Doshi's doubts.
Despite different opinions on the Pfizer vaccine's efficacy, both experts said that Pfizer should reveal the raw data for peer review to better analyze the vaccine, so as to exclude the influence of any environmental or other elements on the efficacy.
https://www.bmj.com/content/bmj/371/bmj.m4058.full.pdf
为了临床试验的透明公正，Moderna 发表了phase 3 covid vaccine clinial trial protocol. 见链接
主要的take home message
primary endpoint: covid infection rate between placebo and vaccine groups
所有的trial, pfizer, AZ, Moderna的 primary endpoint 都 不是比较住院率，疾病严重程度。而是有任何轻微症状 （咳嗽，头痛，发烧，腹泻，e t c) 的受试者中核酸检测阳性的人数和比例。所以真正完全没有任何症状的人不进入统计也是非常合理的设计，因为有无症状是疫苗是否有效的判断。因为疫苗不会让人不接触病毒 （有可能你的鼻粘膜上有病毒的核算片段，不是活病毒），而是打了疫苗以后可以令病毒无法在你接触以后保持活性和扩增，进而染病。
中国的 两个covid 的疫苗trial 既没有release protocol 也没有公布详细得数据。我不认为是全员筛检病毒核酸，因为在疫情泛滥的地区做trial 这样做不合理。尤其是巴西也在同时做很多其他家的trial. 我们就假设中国疫苗trial protocol 也是类似的。测试有任何轻微症状的受试者中核酸阳性率。最新的公布数据是有效率 （有症状的核酸阳性率）50%。
首先各种疫苗的clinical trial 的 primary endpoint assessment,从来都不是通过测试所有参与trial人群筛查核酸阳性率得到疫苗的有效率， 而是计算存在症状+核酸阳性的感染人数。被公认的50%有效率标准也是基于这个计算率。
covid疫苗的设计和有效率计算遵从公认的疫苗大规模人群实验的原理，没有问题。
这个可以引入已经开展很久的流感疫苗trial 常规做法和有效率 （efficacy) 的定义和计算方法。被接受的 positive case definition
A Study to Evaluate the Efficacy of GSK Biologicals' Influenza Vaccine in Childrenhttps://www.clinicaltrials.gov/ct2/show/NCT01218308
To confirm influenza A and/or B disease, （条件1）a positive reverse transcriptase polymerase chain reaction (RT-PCR) result for influenza A or B virus from a nose and throat swab obtained concurrently with an influenza like illness (ILI) （条件2）was required. ILI was defined as the presence of an oral or axillary temperature ≥ 37.8 degrees Celsius (°C) in the presence of at least one of the following symptoms on the same day: cough, sore throat, runny nose or nasal congestion.
19% or 95%? US expert challenges Pfizer vaccine’s efficacy, triggers debates in China By Leng Shumei and Bai YunyiPublished: Jan 13, 2021 10:55 PM
An illustration shows vials of a COVID-19 vaccine and syringes with the logos of US pharmaceutical company Pfizer and German partner BioNTech. Photo: AFP
Questions raised by Peter Doshi, an assistant professor of pharmaceutical health services research at the University of Maryland School of Pharmacy, have triggered heated discussion on Chinese social media.
Doshi's opinion was disputed among Chinese vaccine experts interviewed by the Global Times. Many experts agreed that Pfizer should provide more raw data for peer review but said Doshi's methodology may be flawed.
Chinese experts also noted that disputes often arise from clinical trial data of COVID-19 vaccines, given the difference in participant groups and how different pharmaceutical companies define COVID-19 symptoms differently. They said that the efficacy rate of a vaccine should be evaluated in a more comprehensive manner, citing Sinovac as an example.
Sinovac vaccine received a 50.3-percent efficacy rate on Wednesday, based on phase III clinical trials in Brazil. But the figure was largely due to the fact that participants in Brazil are all medical workers with a very high infection risk, experts said.
Doshi on January 4 released an article on a blog platform under the UK pharmaceutical journal The BMJ, questioning Pfizer's efficacy rate.
Pfizer revealed that it discovered 170 PCR confirmed COVID-19 cases during the phase III clinical trials and 3,410 suspected cases in total.
However, if taking these suspected cases all as confirmed ones, the Pfizer vaccine's efficacy would be dramatically reduced to 19 percent. Even after omitting cases occurring within seven days of vaccination, which should include the majority of symptoms due to short-term vaccine reactogenicity, the efficacy rate remains as low as 29 percent, Doshi said in the article.
Doshi also questioned the standards of how Pfizer excluded cases and the influence of the use of medication on the vaccine's efficacy.
Doshi's comments triggered controversy on Chinese social media platforms, with some Chinese experts doubting Doshi's counting method for efficacy as he takes all participants showing symptoms to be confirmed COVID-19 cases, even if PRC tests show negative results.
"The PRC test is of very high sensitivity, 98.6 percent in the last year. But its sensitivity could be only 5 percent under Doshi's counting method," Zhuang Shilihe, a Guangzhou-based vaccine doctor, told the Global Times on Wednesday.
Zhuang noted that Doshi's doubts may be related to the anti-vaccine movement in the US, noting that Doshi has previously criticized other vaccines like the flu vaccines.
A Beijing-based vaccine expert who required anonymity expressed distrust of Pfizer's efficacy, but he did not directly comment on Doshi's doubts.
Despite different opinions on the Pfizer vaccine's efficacy, both experts said that Pfizer should reveal the raw data for peer review to better analyze the vaccine, so as to exclude the influence of any environmental or other elements on the efficacy.
https://www.globaltimes.cn/page/202101/1212746.shtml
来看看所谓全体筛查的方法是否可行： 首先如果筛查核酸，timing非常重要。在接种以后，数万人来自不同地方，不可能每隔7-14天去做一次鼻咽拭子的核酸测试，因为整个trial 时长跨度一般到达6-12个月。实验参与人数众多，持续性的反复核酸检测的设计并不实际。当然在中国可以弄强制性的把受试者都反复的测，哪怕在从前中国做的流感疫苗trial也没有使用全员受试者每隔7-14天反复测流感核酸。如果你在整个实验结束的时候 （6个月，12个月）只筛查一次，那么你会漏掉”所谓的所有毫无症状感染了病毒已经痊愈的人“。
再看看全体筛查是否有意义： 疫苗的原理不是预防病毒和人群的接触，而是保护人群在接触了病毒以后不发病。在这个基础上，如果受试者没有任何症状，那么他的鼻咽测试核酸结果有两种可能： 核酸阳性： 说明这个人接触了病毒，病毒的核酸片段存在，不能排除病毒在测试的位置是否是活病毒，但是这个核酸的存在不引起发病。核酸阴性，说明这个人根本没有接触到病毒。那么无论哪个结果，从生物学上都说明这个人不是一个病例。
最后看看有症状测试病毒核酸的方法： 当受试者出现任何症状，测试病毒： 核酸阴性说经他的症状不是covid引起的，核酸阳性，说明症状和病毒感染有因果关系，可以作为一个感染病例。
再看一下有效率的计算：
公认的方法： （对照组有症状感染人数/对照组总人数）=对照组的感染率
（疫苗组有症状感染人数/疫苗组总人数）=疫苗组的感染率
疫苗有效率： （对照组感染率-疫苗组感染率）/对照组感染率*100%
这是流感无症状感染率的数据：
The prevalence of asymptomatic carriage (total absence of symptoms) ranged from 5.2% to 35.5%
这是新冠病毒的无症状感染率数据
The proportion of asymptomatic COVID-19 patients reported here is ranged from 17.9% to 35%
我们可以说说dirty pathogen carrier.但是并不会表现出任何病症。
Estimation of the asymptomatic ratio of novel coronavirus infections (COVID-19)
日本的研究
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7270890/
所以越多的人接种疫苗对公共卫生来说阻断流行的机会越高。这个不能discount 疫苗有效率的计算的合理性。
甚至是头疼，头晕，嗓子疼，鼻子堵，肚子不舒服，身上疼痛都算是症状之一。
所以单纯讨论因为无症状而漏掉的病人数，其实意义不大。
许多所谓的无症状者只是症状极轻，发病过程中多少都有一些不舒服。