手機版
ANIL POTTI, Joseph Nevins and their colleagues at Duke University in Durham, North Carolina, garnered widespread attention in 2006. They reported in the New England Journal of Medicine that they could predict the course of a patient’s lung cancer using devices called expression arrays, which log the activity patterns of thousands of genes in a sample of tissue as a colourful picture (see above). A few months later, they wrote in Nature Medicine that they had developed a similar technique which used gene expression in laboratory cultures of cancer cells, known as cell lines, to predict which chemotherapy would be most effective for an individual patient suffering from lung, breast or ovarian cancer.
2006年，北卡羅來納州達勒姆市杜克大學的阿尼爾•波提、約瑟夫•奈文斯及其同事受到廣泛關注。他們在《新英格蘭醫學雜誌》shangfabiaowenzhang，shuoshikeyishiyongyizhongjiaozuobiaodaxuliedeshouduanlaiyucefeiaibingrendeaibianguocheng，zhezhongfangfakeyibazuzhiyangbenzhongchengqianzhongjiyindehuodongxingtaijiluchengyizhangcaisetupian（如上圖）。幾個月後，他們又在《自然醫學》上報告說，他們已經研發出一種同類的技術，可以使用實驗室培養的癌細胞（即細胞係）中的基因表達，來預測對於一個肺癌、乳腺癌或卵巢癌患者來說，哪一種化療方法會最有效。
At the time, this work looked like a tremendous advance for personalised medicine—the idea that understanding the molecular specifics of an individual’s illness will lead to a tailored treatment. The papers drew adulation from other workers in the field, and many newspapers, including this one (see article), wrote about them. The team then started to organise a set of clinical trials of personalised treatments for lung and breast cancer. Unbeknown to most people in the field, however, within a few weeks of the publication of the Nature Medicine paper a group of biostatisticians at the MD Anderson Cancer Centre in Houston, led by Keith Baggerly and Kevin Coombes, had begun to find serious flaws in the work.
當時，這一研究好象是個性化醫療的重大突破——想想看，隻要我們了解一個人的病症在分子層麵有哪些特點，就可以對之進行針對性的治療，這是多麼聰明的想法！因此，這些論文從同一領域的其他研究者那裏獲得了大量讚譽，包括《經濟學人》zaineidexuduobaozhizazhidouzaiwenyuyibaodao。zhegeyanjiuxiaozujiezhekaishizuzhiyixiliedelinchuangshiyan，weifeiaiheruxianaibingrentigonggexinghuadezhiliao。buguo，zhegelingyulideduoshurendoubuzhidao，zai《自然醫學》上那篇論文發表後短短幾周內，休斯頓安德森癌症中心的一群生物統計學家就在基思•拜格裏和凱文•庫姆斯的領導下，開始在上述研究中發現嚴重的缺陷。
Dr Baggerly and Dr Coombes had been trying to reproduce Dr Potti’s results at the request of clinical researchers at the Anderson centre who wished to use the new technique. When they first encountered problems, they followed normal procedures by asking Dr Potti, who had been in charge of the day-to-day research, and Dr Nevins, who was Dr Potti’s supervisor, for the raw data on which the published analysis was based—and also for further details about the team’s methods, so that they could try to replicate the original findings.
拜bai格ge裏li博bo士shi和he庫ku姆mu斯si博bo士shi本ben來lai是shi因yin應ying安an德de森sen中zhong心xin的de臨lin床chuang研yan究jiu員yuan想xiang要yao使shi用yong這zhe項xiang新xin技ji術shu的de需xu要yao，而er準zhun備bei再zai現xian波bo提ti博bo士shi的de研yan究jiu結jie果guo的de。最zui初chu遇yu到dao困kun難nan時shi，他ta們men按an照zhao常chang規gui，請qing負fu責ze日ri常chang研yan究jiu工gong作zuo的de波bo提ti博bo士shi以yi及ji他ta的de主zhu管guan奈nai文wen斯si博bo士shi提ti供gong他ta們men所suo發fa表biao的de論lun文wen原yuan始shi數shu據ju，並bing進jin一yi步bu詳xiang細xi詢xun問wen他ta們men的de研yan究jiu方fang法fa，以yi便bian能neng夠gou複fu製zhi他ta們men原yuan來lai的de發fa現xian。
A can of worms問題多多
Dr Potti and Dr Nevins answered the queries and publicly corrected several errors, but Dr Baggerly and Dr Coombes still found the methods’ predictions were little better than chance. Furthermore, the list of problems they uncovered continued to grow. For example, they saw that in one of their papers Dr Potti and his colleagues had mislabelled the cell lines they used to derive their chemotherapy prediction model, describing those that were sensitive as resistant, and vice versa. This meant that even if the predictive method the team at Duke were describing did work, which Dr Baggerly and Dr Coombes now seriously doubted, patients whose doctors relied on this paper would end up being given a drug they were less likely to benefit from instead of more likely.
波bo提ti博bo士shi和he奈nai文wen斯si博bo士shi回hui答da了le他ta們men的de問wen題ti，還hai公gong開kai更geng正zheng了le幾ji個ge錯cuo誤wu，但dan是shi拜bai格ge裏li博bo士shi和he庫ku姆mu斯si博bo士shi還hai是shi發fa現xian，他ta們men的de所suo謂wei預yu測ce方fang法fa其qi實shi跟gen碰peng運yun氣qi差cha不bu多duo。而er且qie，新xin的de問wen題ti還hai在zai接jie踵zhong而er至zhi。比bi如ru，他ta們men發fa現xian在zai一yi篇pian論lun文wen中zhong，波bo提ti博bo士shi及ji其qi同tong事shi把ba他ta們men用yong來lai推tui導dao化hua療liao預yu測ce模mo型xing的de細xi胞bao係xi標biao錯cuo了le，把ba敏min感gan性xing的de標biao成cheng了le有you抗kang性xing的de，又you把ba有you抗kang性xing的de標biao成cheng了le敏min感gan性xing的de。這zhe也ye就jiu是shi說shuo，即ji使shi杜du克ke大da學xue研yan究jiu小xiao組zu所suo描miao述shu的de這zhe個ge預yu測ce方fang法fa是shi可ke行xing的de——當然，拜格裏博士和庫姆斯博士現在已經表示嚴重懷疑——假如醫生依賴這篇論文的話，就會給病人開錯藥，結果病人獲益的程度可能不但不會增加，反而會降低。
Another alleged error the researchers at the Anderson centre discovered was a mismatch in a table that compared genes to gene-expression data. The list of genes was shifted with respect to the expression data, so that the one did not correspond with the other. On top of that, the numbers and names of cell lines used to generate the data were not consistent. In one instance, the researchers at Duke even claimed that their work made biological sense based on the presence of a gene, called ERCC1, that is not represented on the expression array used in the team’s experiments.
安德森中心的研究員說他們還發現了一個錯誤，就是用來比較基因和基因表達數據的表格錯位了——基ji因yin列lie相xiang對dui於yu表biao達da數shu據ju出chu現xian了le位wei移yi，結jie果guo彼bi此ci不bu相xiang對dui應ying。此ci外wai，用yong來lai得de出chu數shu據ju的de細xi胞bao係xi的de數shu量liang和he名ming稱cheng前qian後hou也ye不bu一yi致zhi。有you一yi個ge地di方fang，杜du克ke大da學xue的de研yan究jiu員yuan甚shen至zhi聲sheng稱cheng由you於yu一yi個ge叫jiao做zuoERCC1的基因的存在，證明他們的研究完全合乎生物科學，可是這個基因在他們實驗所使用的表達係列裏根本就沒有出現。
Even with all these alleged errors, the controversy might have been relegated to an arcane debate in the scientific literature if the team at Duke had not chosen, within a few months of the papers’ publication (and at the time questions were being raised about the data’s quality) to launch three clinical trials based on their work. Dr Potti and his colleagues also planned to use their gene-expression data to guide therapeutic choices in a lung-cancer trial paid for by America’s National Cancer Institute (NCI). That led Lisa McShane, a biostatistician at the NCI who was already concerned about Dr Potti’s results, to try to replicate the work. She had no better luck than Dr Baggerly and Dr Coombes. The more questions she asked, the less concrete the Duke methods appeared.
雖然存在這些有爭議的錯誤，如果杜克大學的研究小組沒有選擇在論文發表後幾個月內（而且當時關於數據質量存在的問題已經開始浮出水麵），jiuyitamendeyanjiuweijichukaishisanxianglinchuangshiyan，zhechanglunzhanbenlaikenenghuijiangjiweikexuewenxianlihuisenandongdezhengbian。botiboshijiqitongshihaijihuabatamendejiyinbiaodashujuyonglaizaiyixiangyoumeiguoguojiaaizhengyanjiusuo（NCI）出資進行的肺癌試驗中指導治療方法的選擇，這就促使原本就對波提博士的研究結果持保留意見的NCI生物統計學家麗薩•麥克申也開始嚐試複製他們的工作。結果跟拜格裏博士和庫姆斯博士一樣，她提出的問題越多，杜克大學的研究方法就顯得越不靠譜。
In light of all this, the NCI expressed its concern about what was going on to Duke University’s administrators. In October 2009, officials from the university arranged for an external review of the work of Dr Potti and Dr Nevins, and temporarily halted the three trials. The review committee, however, had access only to material supplied by the researchers themselves, and was not presented with either the NCI’s exact concerns or the problems discovered by the team at the Anderson centre. The committee found no problems, and the three trials began enrolling patients again in February 2010.
有鑒於此，NCI向杜克大學的管理層就該項研究的實際情形表示了擔憂。2009年10yue，dukedaxuedeguanyuanduibotiboshihenaiwensiboshideyanjiugongzuojinxinglewaibushenhe，bingzanshizhongzhilesanxianglinchuangshiyan。keshishenzhaweiyuanhuizhishoudaoleyanjiufangzijisuotigongdecailiao，tamenjibuzhidaoNCI的保留意見細節，也不清楚安德森中心研究人員所發現的問題。結果委員會沒發現任何問題，三項臨床試驗也於2010年2月重新開始招收病人。
Finally, in July 2010, matters unravelled when the Cancer Letter reported that Dr Potti had lied in numerous documents and grant applications. He falsely claimed to have been a Rhodes Scholar in Australia (a curious claim in any case, since Rhodes scholars only attend Oxford University). Dr Baggerly’s observation at the time was, “I find it ironic that we have been yelling for three years about the science, which has the potential to be very damaging to patients, but that was not what has started things rolling.”
一直到2010年7月，《癌症通訊》報道說，波提博士在若幹文件和撥款申請中造了假，謊稱自己是澳大利亞的羅茲獎學者（這一聲明非常稀奇，因為羅茲獎學者是隻能上牛津大學的），事情才開始敗露。拜格裏博士當時評論說：“我覺得特別具有諷刺意味的是，我們為了他們的研究大喊大叫了三年，因為可能會嚴重損害到病人，可那並沒能讓他們停下來。”
A bigger can?還有更多的問題？
By the end of 2010, Dr Potti had resigned from Duke, the university had stopped the three trials for good, scientists from elsewhere had claimed that Dr Potti had stolen their data for inclusion in his paper in the New England Journal, and officials at Duke had started the process of retracting three prominent papers, including the one in Nature Medicine. (The paper in the New England Journal, not one of these three, was also retracted, in March of this year.) At this point, the NCI and officials at Duke asked the Institute of Medicine, a board of experts that advises the American government, to investigate. Since then, a committee of the institute, appointed for the task, has been trying to find out what was happening at Duke that allowed the problems to continue undetected for so long, and to recommend minimum standards that must be met before this sort of work can be used to guide clinical trials in the future.
2010年底，波提博士離開了杜克大學，而杜克大學也徹底終止了那三項臨床試驗。有其它地方的科學家宣稱，波提博士在《新英格蘭醫學雜誌》上的那篇論文中盜用了他們的數據，於是杜克大學的官員啟動程序，開始撤回連同《自然醫學》上那篇文章在內的三篇重要論文。（《新英格蘭醫學雜誌》上的那篇論文不算在這三篇之內，不過也已於今年3月撤回。）與此同時，NCIhedukedaxuedeguanyuanyaoqingweimeiguozhengfutigongzhuanjiazixundeyixuexiehuizheshoujinxingtiaozha。congnashikaishi，gaixiehuizhuanmenweicichenglideweiyuanhuiyizhizaiyanjiu，zhexiewentiweishenmehuizaidukedaxuechixucunzaizhemejiuermeirenfaxian，yibianjiujianglaibacileiyanjiugongzuoyongyuzhidaolinchuangshiyanqianbixumanzudezuidibiaozhuntichujianyi。
At the committee’s first meeting, in December 2010, Dr McShane stunned observers by revealing her previously unpublished investigation of the Duke work. Subsequently, the committee’s members interviewed Dr Baggerly about the problems he had encountered trying to sort the data. He noted that in addition to a lack of unfettered access to the computer code and consistent raw data on which the work was based, journals that had readily published Dr Potti’s papers were reluctant to publish his letters critical of the work. Nature Medicine published one letter, with a rebuttal from the team at Duke, but rejected further comments when problems continued. Other journals that had carried subsequent high-profile papers from Dr Potti behaved in similar ways. (Dr Baggerly and Dr Coombes did not approach the New England Journal because, they say, they “never could sort that work enough to make critical comments to the journal”.) Eventually, the two researchers resorted to publishing their criticisms in a statistical journal, which would be unlikely to reach the same audience as a medical journal.
在該委員會於2010年12月yue召zhao開kai的de第di一yi次ci會hui議yi上shang，麥mai克ke申shen博bo士shi公gong布bu了le她ta之zhi前qian就jiu杜du克ke大da學xue的de研yan究jiu所suo做zuo的de未wei公gong開kai發fa表biao的de調tiao查zha，結jie果guo令ling觀guan察cha員yuan大da吃chi一yi驚jing。隨sui後hou，委wei員yuan會hui的de成cheng員yuan就jiu拜bai格ge裏li博bo士shi梳shu理li數shu據ju時shi所suo遇yu到dao的de問wen題ti跟gen他ta進jin行xing了le談tan話hua。拜bai格ge裏li博bo士shi說shuo，除chu了le無wu法fa自zi由you取qu得de作zuo為wei該gai項xiang研yan究jiu基ji礎chu的de計ji算suan機ji編bian碼ma和he前qian後hou一yi致zhi的de原yuan始shi數shu據ju外wai，那na些xie發fa布bu了le波bo提ti博bo士shi論lun文wen的de雜za誌zhi都dou不bu肯ken發fa表biao他ta批pi評ping該gai項xiang研yan究jiu的de信xin函han。《自然醫學》發(fa)表(biao)了(le)一(yi)封(feng)他(ta)的(de)信(xin)，連(lian)同(tong)杜(du)克(ke)大(da)學(xue)研(yan)究(jiu)小(xiao)組(zu)的(de)辯(bian)駁(bo)，但(dan)是(shi)在(zai)問(wen)題(ti)持(chi)續(xu)得(de)不(bu)到(dao)解(jie)決(jue)時(shi)，卻(que)拒(ju)絕(jue)發(fa)表(biao)更(geng)多(duo)的(de)評(ping)論(lun)。其(qi)它(ta)後(hou)來(lai)發(fa)表(biao)了(le)波(bo)提(ti)博(bo)士(shi)高(gao)調(tiao)論(lun)文(wen)的(de)刊(kan)物(wu)也(ye)一(yi)樣(yang)。（拜格裏博士和庫姆斯博士沒有跟《新英格蘭醫學雜誌》聯係過，因為他們“一直無法徹底理清那項研究，因而無法向該雜誌提出批評意見”。）最終，這兩位研究員隻得把他們的批評發表在一個統計刊物上，其讀者群自然跟醫學雜誌的不一樣。
Two subsequent sessions of the committee have included Duke’s point of view. At one of these, in March 2011, Dr Nevins admitted that some of the data in the papers had been “corrupted”. He continued, though, to claim ignorance of the problems identified by Dr Baggerly and Dr Coombes until the Rhodes scandal broke, and to support the overall methods used in the papers—though he could not explain why he had not detected the problems even when alerted to anomalies.
該委員會接下來的兩次會議聽取了杜克大學的意見。在2011年3月的那次會議上，奈文斯博士承認這些論文中有些數據“存在缺陷”。不bu過guo，他ta聲sheng稱cheng在zai羅luo茲zi醜chou聞wen發fa生sheng前qian，他ta對dui於yu拜bai格ge裏li博bo士shi和he庫ku姆mu斯si博bo士shi所suo發fa現xian的de問wen題ti一yi無wu所suo知zhi，而er且qie他ta對dui於yu論lun文wen中zhong所suo使shi用yong的de總zong體ti方fang法fa表biao示shi支zhi持chi，雖sui然ran他ta無wu法fa解jie釋shi為wei什shen麼me當dang有you人ren指zhi出chu一yi些xie反fan常chang情qing況kuang時shi，他ta沒mei能neng及ji時shi發fa覺jiao問wen題ti。
At its fourth, and most recent meeting, on August 22nd, the committee questioned eight scientists and administrators from Duke. Rob Califf, a vice-chancellor in charge of clinical research, asserted that what had happened was a case of the “Swiss-cheese effect” in which 15 different things had to go awry to let the problems slip through unheeded. Asked by The Economist to comment on what was happening, he said, “As we evaluated the issues, we had the chance to review our systems and we believe we have identified, and are implementing, an improved approach.”
8月22日，在第四次也是最近的一次會議上，委員會詢問了杜克大學的八位科學家及其管理人員。負責臨床研究的副校長羅勃•卡利夫說，這起事件是典型的“瑞士奶酪效應”，也就是說，有15件不同的事情都出了錯，才導致問題發生而無人察覺。當《經濟學人》就所發生的事請他評論時，他說：“我們在評估這些問題時，也借這個機會審查了我們的係統。我們相信已經找到了、並且正在實施改進的措施。”
The university’s lapses and errors included being slow to deal with potential financial conflicts of interest declared by Dr Potti, Dr Nevins and other investigators, including involvement in Expression Analysis Inc and CancerGuide DX, two firms to which the university also had ties. Moreover, Dr Califf and other senior administrators acknowledged that once questions arose about the work, they gave too much weight to Dr Nevins and his judgment. That led them, for example, to withhold Dr Baggerly’s criticisms from the external-review committee in 2009. They also noted that the internal committees responsible for protecting patients and overseeing clinical trials lacked the expertise to review the complex, statistics-heavy methods and data produced by experiments involving gene expression.
杜克大學的過失和錯誤包括：在處理波提博士、奈文斯博士及其他研究員所公開的潛在財務利益衝突時反應過慢，這其中包括當事人在表達分析公司（Expression Analysis Inc）和癌症指導診斷公司（CancerGuide DX）中(zhong)的(de)介(jie)入(ru)，而(er)這(zhe)兩(liang)間(jian)公(gong)司(si)跟(gen)杜(du)克(ke)大(da)學(xue)也(ye)有(you)關(guan)聯(lian)。此(ci)外(wai)，卡(ka)利(li)夫(fu)博(bo)士(shi)及(ji)其(qi)他(ta)高(gao)級(ji)管(guan)理(li)人(ren)員(yuan)也(ye)承(cheng)認(ren)，研(yan)究(jiu)工(gong)作(zuo)出(chu)問(wen)題(ti)以(yi)後(hou)，他(ta)們(men)過(guo)於(yu)倚(yi)賴(lai)奈(nai)文(wen)斯(si)博(bo)士(shi)和(he)他(ta)的(de)判(pan)斷(duan)，以(yi)至(zhi)於(yu)在(zai)2009年(nian)舉(ju)行(xing)那(na)次(ci)外(wai)部(bu)審(shen)查(zha)時(shi)，沒(mei)有(you)把(ba)拜(bai)格(ge)裏(li)博(bo)士(shi)的(de)批(pi)評(ping)意(yi)見(jian)告(gao)訴(su)審(shen)查(zha)委(wei)員(yuan)會(hui)。他(ta)們(men)也(ye)注(zhu)意(yi)到(dao)，大(da)學(xue)內(nei)部(bu)負(fu)責(ze)保(bao)護(hu)病(bing)人(ren)和(he)監(jian)督(du)臨(lin)床(chuang)試(shi)驗(yan)的(de)委(wei)員(yuan)會(hui)缺(que)乏(fa)專(zhuan)業(ye)知(zhi)識(shi)，無(wu)法(fa)審(shen)查(zha)基(ji)因(yin)表(biao)達(da)試(shi)驗(yan)中(zhong)所(suo)使(shi)用(yong)的(de)複(fu)雜(za)統(tong)計(ji)方(fang)法(fa)及(ji)其(qi)所(suo)得(de)出(chu)的(de)數(shu)據(ju)。
That is a theme the investigating committee has heard repeatedly. The process of peer review relies (as it always has done) on the goodwill of workers in the field, who have jobs of their own and frequently cannot spend the time needed to check other people’s papers in a suitably thorough manner. (Dr McShane estimates she spent 300-400 hours reviewing the Duke work, while Drs Baggerly and Coombes estimate they have spent nearly 2,000 hours.) Moreover, the methods sections of papers are supposed to provide enough information for others to replicate an experiment, but often do not. Dodgy work will out eventually, as it is found not to fit in with other, more reliable discoveries. But that all takes time and money.
這是一個調查委員會反複聽到的主題。同行審查這一程序依靠的是同一領域其他同行的善意（一直以來都是如此），但是這些同行也都有他們自己的工作，通常沒有足夠的時間來徹底仔細地檢查別人的論文。（麥克申博士估計，她花了300-400小時檢查杜克大學的研究；拜格裏博士和庫姆斯博士估計，他們花了將近2,000小時。）此ci外wai，論lun文wen的de方fang法fa部bu分fen本ben來lai應ying該gai為wei其qi他ta人ren複fu製zhi實shi驗yan提ti供gong足zu夠gou的de信xin息xi，但dan通tong常chang情qing況kuang都dou不bu是shi這zhe樣yang。騙pian人ren的de研yan究jiu如ru果guo跟gen其qi它ta更geng可ke靠kao的de發fa現xian相xiang違wei，最zui終zhong會hui被bei拆chai穿chuan，但dan這zhe一yi切qie都dou需xu要yao時shi間jian和he費fei用yong。
The Institute of Medicine expects to complete its report, and its recommendations, in the middle of next year. In the meantime, more retractions are coming, according to Dr Califf. The results of a misconduct investigation are expected in the next few months and legal suits from patients who believe they were recruited into clinical trials under false pretences will probably follow.
醫yi學xue協xie會hui計ji劃hua明ming年nian年nian中zhong完wan成cheng報bao告gao及ji其qi建jian議yi。與yu此ci同tong時shi，按an照zhao卡ka利li夫fu博bo士shi的de說shuo法fa，還hai會hui有you更geng多duo的de論lun文wen被bei撤che回hui。在zai接jie下xia來lai的de幾ji個ge月yue裏li，瀆du職zhi調tiao查zha會hui見jian分fen曉xiao，隨sui之zhi而er來lai的de可ke能neng是shi因yin為wei虛xu假jia陳chen述shu而er報bao名ming參can加jia臨lin床chuang試shi驗yan的de病bing人ren提ti起qi法fa律lv訴su訟song。
The whole thing, then, is a mess. Who will carry the can remains to be seen. But the episode does serve as a timely reminder of one thing that is sometimes forgotten. Scientists are human, too.
整件事情一團糟。最後誰會成為炮灰還不知道，不過這些情節提醒我們記住一個常常被忽略的事實——科學家也是人。
Correction: This article originally stated that by the end of 2010 officials at Duke University began the process of retracting five papers. That should have been three papers. This was corrected on September 8th.
更正：本文原來說，截止到2010年底，杜克大學的官員啟動程序，開始撤回五篇論文，實際為三篇，故於9月8日更正。
原文鏈接：http://www.economist.com/node/21528593
相關閱讀：東北農業大學實驗室感染事故引發的安全思考
