一种定量评估化学物质致癌性的新方法——比值评估法

A NEW APPROACH TO PREDICT CARCINOGENICITY OF CHEMICALS QUANTITATIVELY——RATIO ESTIMATE METHOD

作者提出一种在诱变试验中定量评估化学物质致癌性的新方法——比值评估法。该法应用有关概率公式,结合各试验的灵敏度和特异度,用比值简化运算,求得化学物在结果群条件下为致癌物的概率,此方法的优点在于只需了解P(C^+)的范围,而不必先确定其确切值,并且只作较简便的运算,就能得到P(C^+/M)的估计值及误差。此法在毒理测试结果评价中有一定实用价值。

Based upon the results of battery ofmutagenesis tests, a new approach, ratioestimate method (REM), for predictingcarcinogenicity of chemicals quantitativelyis described. If the tests arc composed of nmutually independent experiments, a re-sult's set is constituted:M={M_1, M_2,…, M_n}.According to the formulas of probability, wehave: P(C^+/M)/P(C^-/M)=[P(C^+)/P(C^-)].[multiply from i=1 to n(P(Mi/C^+))/multiply from i=1 to n(P(Mi/C^-)),whereP(C^+/M)/P(C^-/M)is the ratio of the proba-bility of a carcinogen to a non-carcinogenwith the result's set of M. P(C^+)and P(C^-)signify the probability of a carcinogen anda noncarcinogen. P(M_i/C^+)and P(M_i/C^-)represent the probability with which theresult M_i can be obtained to a carcinogenor a noncarcinogen. Let1/α=p(c^+)/P(C^-),=P(C^+/M)/P(C^-/M), γ=multiply from i=1 to n(P(M_i/C^+))/multiply from i=1 to nP(M_i/C^-), then β=γ/α. The value γ can bederived from the sensitivity and specificityof mutagensis tests. It is difficult to get theexact value of a but one can give its suitablerange. For example, let a≤α≤b, thenγ/γ+b≤P(C^+/M≤γ/γ+a Thus, the mean value ?(C^+/M)=γ/2((1/γ+b)+(1/γ+a)) could be taken as anestimated probability to predict the carcino-genicity. The application of REM to predictthe carcinogenicity of a carcinogenic and anon-carcinogenic polycyclic aromatic hydro-carbon was illustrated. The error range ofthe value estimated by REM was also dis-cussed in detail. The major advantange of REM is that itonly demands to give an experiential rangeof P(C^-)/P(C^+)without pre-determining itsexact value and the calculating process issimple and feasible to the toxicologists.