Document 7Oo8q0GwdnKKk3Jz3xNb20po8

MALIGNANCIES IN DANISH THOROTRAST PATIENTS* MOCENS FABER Finsen Laboratory, Strandboulevard49, Copenhagen, Denmark Abstract-The Danish Thorotrast follow-up has been running since 1949.As of 1973, 1012 cases were known with 391 still living. The malignancies are classified as Thorotrast inducible and other tumors. Thorotrast inducible were 11 hemangiocndotheliomas, 9 cholangiocarcinomas, 12 hepatomas and 11 leucaemias. In addition, 52 other cancers were observed. The 3 types of liver tumors (hemangioendotheliomas, cholangiocarcinomas, and hepatomas) showed a similar number of cases, latency periods, rises in cumulative incidence with time, and radiation doses to the liver. No bone tumors were observed. The 11 cases of respiratory cancer were all in males with 4 having latency periods under ISyr, which does not support a radiation induction. However, the high ratio of anaplastic lung tumorslepidermoid lung tumors, agrees with findings in the uranium miners, and could indicate some influencefrom radiation. It is uncertain whether the two cases of pleural mesothelioma may have been induced by Thorotrast because there is no evidence for significant irradiation of the pleura. It is concluded that a larger population or a longer follow-up is necessary to exclude or include bone sarcomas and lung tumors as being due to Thorotrast. INTRODUCTION HE DAN~STHhorotrast patients have been ider observation since 1949 and surveys of le status of the study has been made at ,tervals (Faber 1%2; 1%7; 1973a). The itients were injected intravascularly beJeen 1935 and 1946 for arteriography of :urological disorders, except for one case jected for a haemangioma on the foot. MATERIAL Apart from an increased incidence of brain mors, for which the Thorotrast was somenes injected, the population was considered being equivalent to the general Danish )pulation, so' that the figures from the anish cancer registry ((364; 1965; 1969) mld be used when expected tumor indences were calculated. This was, furtherore, an absolute requirement as it was imjssible to collect a control group which *Dr.Faber was unable to attend, but advance )pies of his manuscript were distributed at the mposium. would match exactly in sex ratio, age distribution, and date of entrance into the study. It is important in a study, which now covers close to 40yr of observation, that the shift in cancer incidence with time is taken into consideration in the calculations. A reasonable equivalent of the possible carcinogenic factors connected with the disease for which the Thorotrast was given is also of importance. These two points practically exclude the possibility for collecting a control group in Denmark because all cases with the same diseases had their normal diagnostic studies at the hospital where the Thorotrast was used. Thorotrast injected persons could be expected to deviate from the normal population by an increased risk of exposure to carcinogenic drugs. The group contains a number of patients with lifelong headaches and concomitant use of many different drugs. Smoking may be predominant, and we know that the suicide rate is high. About 10-20% of the patients had epilipsia. Finally, it has to be remembered that a few of the patients were 153 given X-ray therapy for some of the diseases for which Thorotrast was injected. The follow-up as of 1973, is summarized in Table 1. The observation time of the patients in relation to injected dose is presented for males and females in Tables 2 and 3. Table 4 registers the malignant tumors. The estimates of expected malignancies have been calculated from the size and age distribution of the remaining population, year by year. The number of cancers and leucaemias had now grown to 95. They can, from previous experiences, be split into two groups. Those where the dependency on Thorotrast is recognized and the others. THOROTRAST INDUCIBLE TUMORS The dependent malignancies are liver tumors and leucaemias. Table 1. Iknish Thomtrasf injected patients Dud tiviry MI 210 mo 181 621 391 Told 571 441 1012 Three main groups of liver tumors found; the haemangioendotheliom cholangiocarcinomas, and the hep (hepatic cell carcinomas). The hemangioendothelioma has r been shown, both in human ex 1 (Fa62), and in animal experiments be a type of tumor intimately conn I Thorotrast deposits. The primary mostly located in the liver in man, but in the spleen in the rabbit. 10' A knowledge of in relation to the the pattern time after of appear& Thorotrast 'b 1 jection is useful when the Thorotrast &., ducible tumors are compared with the 0- tumors found in the Thorotrast patients. difference in behaviour can be seen in Fw.1. The cumulative incidence of those turnom where, so far, no causal relation to I Thorotrast has been demonstrated, rises by 8 straight line starting immediately after & I injection. 1 1The Thorotrast inducible tumors, howem, show a latency period without the a .* of any tumor. This latency period i Table 2. l h n i r h Tkomtmst palienfa according IO h e and rime of obsemotion-mdea Yr M m l 1C19ml M 2 9 m l W 3 9 m l M m l Wml unknown alluser CI 4 45 43 13 7 0 22 I 3 4 2 2 3 7 I 4 17 3 3I I 27 4 14 I 28 5 63 2 II 6 22 I 27 75 05 8I 3 1 I6 9 33 I I I9 IO 3 5 I 2 I 2 14 I1 I 4 32 I II 12 2 3 2 7 13 4 2 2 8 14 6 2 I 9 I5 5 I 6 16 2 2 4 17 3 I I 5 I8 4 I I I7 m19 I II 42 I I I I 6 7 2n1 32 I3 I 28 4 23 2 2 I I 6 24 2 3 I 2 8 73 I 34 2n6 33 I I8 IO I4 I 4 3 5 37 m I 15 12 4 2 34 29 4 21 13 2 1 2 43 30 3 I5 5 4 2 I 30 31 4 I2 5 I I I 24 32 2 12 6 3 I 33 5 7 3 I I 5 4 3m0 34 5 4 I I II 35 5 3 I I I I1 36 3 0 I4 37 38 I I MUXS 571 4 i II i4 1 t I 1 i MOGENS FABER Yr 0-1 2 3 4 5 6 7 8 9 10 I1 I2 13 14 I5 16 17 18 19 20 21 22 23 24 25 1n 2m-8 30 31 32 33 34 35 36 37 38 Au c u e s Table 3. W i s h Thomtrart palimts according to dose u d time of ob$emotion--femakI b 9 m l 1O-19ml M29ml 30-39ml M a r l W d unLnown 1 33 31 9 4 3 16 II 6I II 2 3II I 2 I II 3 3I 2 2I I 43 1 2 I 3 2I 1 3I 23 I I 3 I 1I I 2I I I 32 I I I 1 2 I I II 1 I 2 1I 22 32 2 I I3 I I 42 I1 I6 3 4I 3 7 15 13 2 I 2 3 13 9 1 2 I I2 7 1 3 I1 6 I 1 7 I I5 5 5 1 4 4I I2 12I I 84 I 2 1I 1 2 allcue1 91 2 8 6 6 2 10 6 8 3 4 5 6 4 2 3 4 10 1 4 I 4 4 7 6 8 45 36 25 22 28 m 13 4 16 2 1 2 441 155 tween the different malignancies. As was to be expected, the leucaemias had the shortest U u m latency period of about 8yr. The Curve for the rise in cumulative number fobws reasonably close to a straight line. The curve for the primary liver carcinomas pad the one for the haemangioendotheliomas a minimum latency period close to 17 yr. Both rise reasonably rectilinearly with time, 4 io doubtful, when the data in Table 5 @ken into consideration, whether these two curves are different. They are, therefore, combined as a single curve in Fig. 1. The primary liver cell tumors are generally described as being either primary hepatomas I 156 DANISH THOROTRAST MALIGNANCIES FIG. 1. Cumulative number of liver tumors, leucemias, and combined other tumors in the Danish Thorotrast patients by year after Thoro- trast injection. or cholangiocarcinomas on the basis of differences in histology. It is doubtful whether the difference is meaningful, but if this should be the case, it is of interest to know whether the histologic findings reflect a difference in carcinogenic action of the Thorotrast during the prolonged stay in the liver. Both in humans and in rabbit experiments it is evident that the Thorotrast with time is collected in connective tissue scars where most of the granules are localized in macrophages or in multinuclear cells. This connective tissue is the result of repeated episodes of liver cell death with a proliferation of new liver cells in nodules without Thorotrast. In the connective tissue bands we find a large number of surviving bile ducts, apparently unchanged, and surrounded by cells that contain Thorotrast. If during the evolution to this stage there should be a difference in the radiation exposure to the two cell systems, hepatic cells and bile d m cells, this could be reflected in the type of liver tumor found. However, an analysis of the 12 hepatomas, 9 cholangiocarcinomas, and 11 hemangioendotheliomas (Table 5 ) does not give any support for such a separate behaviour with respect to the number of cases, latency period, rise in cumulative in- cidence with time, and a d o s e to the liver calculated from the estimates of Kaul (1%9) and Faber (1973b). In the evaluation of these curves a certain unreliability has to be recognized. As of January 1973, it was 38yr since the first patient was injected with Thorotrast. It was, however, only 27yr since this happened to the last patient. This means that only that part of the curves in Fig. 1 which covers the first 27yr after the injection represents a completed observation. During the next years more cases will probably appear, and they may change the last part of the curves. It is interesting that no cases of leucaemia have appeared later than 23yr after the in- jection, and that the same is the case with the nonleucaemic haematological diseases. This may be a sign that the spatial distribution of the irradiation to the bone marrow has become such that little radiation is received by the sensitive cells. This is the case in rabbit experiments with Thorotrast enriched with T h , but is not very probable in man. The significance of this observation for the Thorotrast patients can not be resolved from this material alone. (YI'EER TUMOIls Bone tumors Due to the translocation of thorium daughters from the liver and spleen deposits to the bone, the bone surfaces receive protracted irradiation (Rowland and Rundo, )973), and it should be possible to induce bone sarcomas in this ways. The incidence of bone sarcomas in Denmark from 1943-1967 bas been unchanged and close to one case per 100,OOO persons/ yr, with only small variations according to sex, and without any marked change in incidence with rising age. The expected number of MOGENS FABER 157 Iy-occurring bone sarcomas can, injected with low doses. The age at ap- w e f o r e , be calculated from the 18,000 pearance corresponds to the one observed in person-yr of observation (9973 male-yr plus the general population. The latency periods 8,372 female-yr). This predicts 0.2 naturally- could indicate that the high incidence was occurring cases expected over the entire purely fortuitous when it is considered that present Danish Thorotrast follow-up. For the four out of eleven cases had a latency period person-yr at risk beyond a reasonable latent shorter than the 15yr to be expected for a peri"od, less than 0.1 naturally-occurring bone radiation induced solid tumor. sarcoma is expected. A support for a radiation genesis is to be The lack of observed cases may be due to a found in the histological typing of the lung , number of factors of which insufficient dose tumors with 5 cases of anaplastic tumors and and insufficient dose rate are of importance. 2 epidermoid tumors. Saccomanno et al. Tbe estimated expectancy for cancer in- (1971) have analysed the type of tumor found duction after incorporation of 224Rain relation in uranium miners and found an increase of to dose is based on observations of German the anaplastic against the epidermoid tumors. pntients where all of the dose was received The pulmonary tumors from the Danish ' during a short period of time after the last %I injection, in contrast to the Thorotrast Thorotrast patients show the same skewness in the histological typing. For both of the cases where the bone dose is being built up populations it must be remembered that such with time. This contrasting mode of dose a division may be erroneous due to the fact W v e r y could give a difference both in the that the histological typing may not be d6beffect relationship and in the duration homogenous throughout the tumor (Clem- a m effective latency period. The conclusion mensen, 1%5). be that a longer observation time is Our material contains two cases of sary to be able to observe the existence histologically verified mesothelioma of the ignificant number of bone sarcomas in a pleura, and one additional case is suspected lation of this size. as being of this type. When taking into ly, a complicated pathophysiological consideration that the expected incidence is n. The incidence of bone sarcomas in virtually zero, this is too large a number. If s been of the same order of magnitude pleural tumors should be due to Thorotrast the Danish cancer registry since 1945. The this might require a subpleural deposit of evaluation of this figure in the elderly can, thorium in lymphatic vessels. Histological bwever, be criticized. In the clinic, bone evidence from both human and injected -coma is a disease of youth and young animals does not support such a deposition. W&ts-not of the elderly. Unless a complete Finally, those patients who present more btological study has been performed, the than one malignancy should be searched for of a misdiagnosis in relation to un- evidence of other Thorotrast induced tumors. SPecified metastases within bone must exist. A survey of the three patients with this behaviour show that two have died from Respiratory tumors classical Thorotrast tumors. The other tumors umors which at present are of great could not be suspected of being directly in- are the 9 in the lung proper and the 2 duced by Thorotrast. eliomas in the pleura (Table 4). The expectation is approximately four CONCLUSION could suggest an increase which As a result of this analysis of the cancers in e to the Thorotrast. the Danish Thorotrast patients, the evaluating these cases as a result of conclusion is clear that liver cancers and iation from thorium daughters, a leucaemias can be induced by Thorotrast. A of points have to be taken into ac- larger population or a longer follow up time is 3 he amount injected does not support needed to resolve the question of whether * se effect relation as most cases were other tumors, such as those of bone or lung, 158 DANISH THOROTRAST MALIGNANCIES .are induced by intravascularly injected Thorotrast REFERENCES Clemmesen J.. 1965, Statistical Studies in Malignant Neoplasms, Vol. I, Kobenhavn: Munks&aard. Clemmesen J., 1964, Statistical Studies in Malignant Neoplasms, Vol. 11, p. 194 Kobenhavn; Munksgaard. Clemmesen J., 1969, Statistical Studies in Malignant Neoplasms, Vol. 111, Kobenhavn: Munksgaard. Clemmesen J.. Fuglsang-Frederiksen V. and Plum C. M.,1974, Lancet. i. 705. Faber M.,1962. in: Some Aspects of Internal Ir- radiation, p. 473 (Edited by Dougherty T. F.. Jee W. S. S.,Mays C. W. and Stover B. J.), Oxford: Pergamon Press. Faber M.,1967, Ann. N.Y. Acad. Sci. 145,843. Faber M.,1973a, "Follow-up of Danish Thorotnut cases'' in: Proceedings of The Third Inter- national Meeting on The Toxicity of Thorotrast, Riso Report No. 294, p. 137. Faber M.,1973b. "Dose effect relations in hepatic carcinogenesis" in: Proceedings of The Third International Meeting on The Toxicity of thomtrast, Riso Report No. 294. p. 308. Johansen C., 1955, Acta path. microbiol. scand. suppl. 105.92. Kaul A., 1969, Zsotopenpraxis 5, 85. Rowland R. E. and Rundo J., 1973, "The Skeletal Dose from =Ra Following Intravascular Ad- ministration of Thorotrast" in: Proceedings of The Third International Meeting on The Toxicity of Thorotrast, Riso Report No. 294, p. 95. Saccomanno G.. Archer V. E., Kuschner M., Saunders R. P. and Klein M.G.,1971, Cancer 27, 515.