Path. Res. Pract. 186, 717-722 (1990)
Increased Content of Chromium and Nickel in Lung Tissues from Patients with Bronchial Carcinoma L. A. Akslen and A. O. Myking Department of Pathology, The Gade Institute, University of Bergen, Bergen, Norway
O. M0rkve and A. Gulsvik Department of Thoracic Medicine, University of Bergen, Bergen, Norway
H. J. Raithel and K. H. Schaller Institute of Occupational and Social Medicine and Policlinic for Occupational Diseases, University of Erlangen, Erlangen, FRG
SUMMARY
Epidemiological studies have shown that occupational exposure to certain chromium and nickel compounds is followed by an increased lung cancer incidence. However, few data exist on the content ofthese metals in lung cancer patients in general. In the present study, central and peripheral lung tissue, bronchial tissue and hilar lymph nodes were collected from 20 patients with bronchial carcinoma and 21 control individuals, and the tissue concentration ofchromium and nickel was measured by use ofatomic absorption analysis. Increased levels of both metals were found in cancer patients as compared to controls. Lung tissue concentration of chromium was two-fold increased, while the bronchial wall content of nickel was three times the level in control individuals. Smokers showed a dose-related increase in the deposition of both chromium and nickel. Furthermore, in cancer patients an inverse relationship between smoking and the tissue level ofchromium in regional lymph nodes was found, possibly indicating a depressive effect on pulmonary clearance mechanisms. Our results emphasize the possible role of small amounts of chromium and nickel as agents in bronchial carcinogenesis, unrelated to occupation and probably related to tobacco smoking.
Introduction Several epidemiological studies have shown that occupational exposure to hexavalent chromium3,5,6, 12, 13 and nickel compounds4, 14, 15 is associated with increased incidence of lung cancer. In addition, very high contents of chromium have been found in lung tissues and other organs from persons with a known and long-lasting exposure9, 10, 21. Mutagenicity of chromium compounds has been well documented, and the interaction with nucleic © 1990 by Gustav Fischet Verlag, Stuttgart
acids may be the basis for its carcinogenic influence l6 • Research on nickel has also shown a wide range of toxicological and carcinogenic effects20 • Although occupational hazards of chromium and nickel are well established, information on the tissue content of these metals in cancer patients is very sparse. However, recent observations indicate that patients with lung cancer may have an increased content of chromium in lung tissue 2, 11. The role of this finding in bronchial carcinogenesis still remains unknown. The aim of the present study was to investigate 0344-0338/90/0186-0717$3.50/0
718 . L. A. Akslen et al.
the content and distribution of chromium and nickel in the lung, bronchial and tumor tissue as well as hilar lymph nodes in a group of unselected lung cancer patients, compared to a control group. Material and Methods
Patients Twenty concomitant patients with bronchial carcinoma were included during a 6 month period in 1986, of these 15 males and 5 females (median age 62 yrs). In four cases (2 males, 2 females) surgical specimens were examined, while the rest were autopsy cases. All patients had been hospitalized in the Department of Thoracic Medicine, Haukeland Hospital, Bergen. Bronchial carcinoma was the main cause of death in the autopsy cases. Altogether 21 control patients were included during the same period, 9 males and 12 females (median age 72 yrs). The main causes of death were ischaemic heart disease (n = 10), cerebrovascular disease (n = 6), chronic respiratory disease (n = 2), aortic stenosis (n = 1) and glioblastoma multiforme (n = 1) without tumor spread. A standardized dissection was performed in all cases. Lung cancer patients and controls were from a limited geographical area in and around Bergen at the west coast of Norway. No production of chromium and nickel is present in this area. A detailed occupational history was obtained for each patient, based on the hospital records and on information from the National Health Service. Occupational exposure to chromium and nickel 19 was graded as none, possible, and probable. Details on smoking habits were available in 17 of 20 cancer cases and 9 of 21 controls, and the quantity is given as grams per week. Histological type of tumor, according to the WHO classification 22 as well as patient survival (months) in fatal cases were registrated.
Tissue Sampling Lung tissue was collected from central and peripheral parts of each lobe, giving a total of 10 specimens from each autopsy case. Instruments free of chromium and nickel were used. In addition, specimens from the central bronchial walls and hilar lymph nodes from both sides were included, together with tumor tissue in cancer cases. In the surgical specimens, sampling from tissue surfaces and resection margins was avoided. Although contamination from surgical instruments cannot be totally excluded, the content of chromium and nickel in these four cases were found to be within the range of the autopsy cases. The specimens, most of which had a dry weight of 20 mg-300 mg, were fixed in absolute alcohol. The analysis of chromium and nickel (ng/g dry weight) was performed at the University of Erlangen (FRG) as described by Raithel et alY. Metal eluation and contamination during specimen storage in alcohol has been analyzed and found to be negligible. The variation coefficient of repeated measurements was 6% for chromium and 9.5% for nickel.
Statistical Methods The median values of chromium and nickel concentration in each tissue type (lung, bronchial, lymphoid, tumor) were calculated for each patient and used in further analysis of differences between cancer cases and control individuals. Differences were compared using the non-parametric Mann-Whitney rank-sum test. Association between different variables is expressed by the non-parametric correlation coefficient Spearman's rho (r), with
corresponding p-values. Probabilities less than 0.05 were considered significant.
Results
There was a marked intraindividual variation in the content qf chromium and nickel in lung tissue (these findings will be presented in a separate publication). Briefly, the mean ratio between maximum and minimum values was as follows: chromium among cancer cases 12.4 (range 2.9-43.3), chromium among controls 8.7 (range 2.3-21.3), nickel among cancer cases 14.1 (range 2.8-67.3) and nickel among controls 11.4 (range 2.6-29.7).
Chromium Concentration The concentration of chromium in lung tissue from cancer patients was more than twice that of controls (Table 1, Fig. 1, P = 0.031). The difference was present among both males (ratio cases/controls 1.5, p = 0.07) and females (ratio cases/controls 3.0, p = 0.19), but the numbers were small for statistical analysis (Table 2). No significant differences were found in bronchial or lymphoid tissues, but cancer cases tended to have somewhat higher concentrations than control individuals. Tumor tissue contained about twice as much chromium as bronchial tissue, but the concentration was clearly lower than in the lung tissue.
LUNG TISSUE
-
~
Cl C1l
.a 12000
•
~
Cl
•
Cl C
z
0
.... 0.05).
The chromium concentration was highest in lymphoid tissue, being about 4.1 times that of lung tissue in cancer cases and 6.1 times among controls. The lowest concentration was found in bronchial tissue, with a bronchiaVlung ratio of 0.19 and 0.24 in cancer cases and controls, respectively. There was no covariation between chromium concentrations of any tissue among control individuals, except a positive correlation between bronchial tissue and lymph nodes (r = 0.47, p < 0.05). In cancer cases, the bronchial tissue concentration was negatively correlated to the content in tumor tissue (r = -0.39, p < 0.05). The lung tissue concentration of chromium was clearly higher in males than females, in cancer cases as well as controls (Table 2). The male/female ratio among controls was 2.4 and 1.2 for cancer patients. However, in lymph nodes and bronchial tissue, females had the highest concentrations in both patient groups.
Table 3. Median concentration of nickel (ng/g dry weight) in various tissue types among cancer patients (n = 20) and control individuals (n = 21)
Tissue type Lung tissue cases controls
Males (N)
Females (N)
P-value l
Lung tissue cases controls
196 165
94 90
507 258
20 21
Bronchial tissue cases controls
416 142
71 36
1312 211
20 20
Lymphoid tissue cases controls
958 1084
477 261
2098 2087
19 21
191
97
395
13
NS NS
Bronchial tissue cases controls
546 (15) 189 (9)
1294 (5) 440 (12)
NS NS
Lymphoid tissue cases controls
12081 (14) 9335 (9)
22000 (5) 15639 (12)
1310 (12)
2020 (5)
Tumor tissue 1
Mann-Whitney rank-sum test (NS denotes p
0.037 NS NS
> 0.05).
p-
value l NS
Tumor tissue
0.026
NS
Ql = Lower quartile; Q3 = Upper quartile, N = Number of cases.
1
Mann-Whitney rank-sum test (NS denotes p
> 0.05).
BRONCH IAL TISSUE
-
10 4
~
01 QI
~
...>. '0
01
10
•
••
•
•
•
3
~ r
01
c:
0
i=
« a::: ~
Z
w u
102
It
I
z
0
2851 (5) 957 (12)
N
value
U
3441 (15) 2253 (9)
Qr
value
z
Table 2. Median concentration of chromium (ng/g dry weight) in various tissue types and specified for both sexes (N = number of cases)
Ql
Median value
Tissue type
I
~ • •
•
I
...J
W
:::.::
u Z
10 1
1
• i
CASES
i
CONTROLS
Fig. 2. Individual log-values of nickel concentration in bronchial tissue from cancer cases and control patients (median values indicated).
720 . L. A. Akslen et al.
tion of nickel was slightly higher in cancer cases, but the difference was not statistically significant (Table 3). However, in lymphoid tissue there was no difference between cancer cases and controls. Tumor tissue showed values comparable to that of lung tissue. Similar to chromium, the nickel concentration was highest in lymphoid tissue, 6.6 times that of lung tissue in control patients and 4.9 among cancer cases. In contrast to chromium, the deposition of nickel in the bronchial walls was about twice that in lung parenchyma among cancer cases, while there was no such difference among controls. The nickel concentration in various tissue types was significantly correlated, in both cancer patients and controls. The strongest correlation in control individuals was lung - bronchial tissue (r = 0.67, p < 0.005) and lung lymphoid tissue (r = 0.67, p < 0.001). In cancer cases, the lung - bronchial correlation (r = 0.78, p < 0.001) and bronchial tissue - lymph node correlation (r = 0.66, p < 0.005) were strong as well. The male cancer cases tended to have a higher lung tissue concentration of nickel than females, without such sex difference among controls (Table 4). In bronchial and lymphoid tissue females had somewhat higher values, in both cancer patients and control individuals, without being statistically significant.
Chromium and Nickel A significant correlation between the tissue concentration of chromium and nickel was found in lung (r = 0.51, p < 0.05) and bronchial tissue (r = 0.43, P < 0.05) among cancer cases, and in bronchial tissue (r = 0.46, p < 0.05) and lymph nodes (r = 0.79, p < 0.001) among control individuals.
Age The median age of cancer patients was 62 yrs, as compared to 72 yrs for control patients (p = 0.0014). Among cancer patients, the median age of males and females were 64 yrs and 58 yrs, respectively. The corresponding figures for control individuals were 72 yrs and 78 yrs. Within each group the age differences were not statistically significant. There was no significant correlation between age and concentration of chromium and nickel in any tissue, neither in cancer patients nor in controls.
Histology There was some variation between different histological tumor types. However, the number of cases in each group was small. The lung tissue concentration of chromium was rather low in the three cases of small cell carcinoma, with a median value of 2279 (ng/g dry weight), compared to 3129 (ng/g) in adenocarcinomas (n = 6) and 3898 (ng/g) in squamous cell carcinomas (n = 8). The nickel concentration in bronchial tissue was high in unclassified carcinoma (2464 ng/g; n = 3), compared to squamous cell carcinoma (510 ng/g) , adenocarcinoma (291 ng/g) and small cell carcinoma (401 ng/g).
Table 4. Median concentration of nickel (ng/g dry weight) in various tissue types and specified for both sexes (N = number of cases Males (N)
Tissue type Lung tissue cases controls Bronchial tissue cases controls
Females (N)
P-value 1
243 (15) 165 (9)
142 (5) 154 (12)
NS NS
401 (15) 71 (9)
431 (5) 159 (11)
NS NS
Lymphoid tissue cases controls
882 (14) 478 (9)
1075 (5) 1377 (12)
NS NS
Tumor tissue
191 (9)
227 (4)
NS
1
Mann-Whitney rank-sum test (NS denotes p > 0.05).
Survival The median survival from the time of cancer diagnosis in all fatal cases (n = 16) was 6 months (range 1-24 months). The concentration of chromium in bronchial tissue was positively correlated to increasing survival time (r = 0.72, p < 0.005), while the concentration of nickel in lung tissue increased somewhat with survival (r = 0.50, p < 0.05). No other statistically significant correlations were found.
Occupational History As judged by a detailed occupational history, four males, three of them having cancer, had a possible or probable exposure to chromium, and one of these cases (no. 15) also had a possible exposure to nickel (Table 5). The lung tissue concentration of chromium among the three cancer cases was 1.6 to 4.0 times that of controls, and the mean value was 1.3 times that of other cancer patients. The one control individual had a chromium concentration of 1.8 times that of other controls. The one cancer patient with a possible exposure to nickel had low tissue concentrations. Table 5. Concentrations of chromium and nickel (ng/g dry weight) in four patients with possible or probable occupational exposure Tissue concentration Patients
Group Exposure l Lung tissue
Bronchial Lymphoid tissue tissue
Chromium Case 2 Case 15 Case 26 Case 41
CA CA CA CO
(+) (+)
6734 2736 3476 3073
1480 282 165 556
21080 9823 13191 15609
Nickel Case 15
CA
(+)
140
62
198
CA = cancer patient, CO = control patient. possible, + = probable.
1
Exposure: (+) =
Chromium and Nickel in Lung Cancer' 721
Smoking Habits Among cancer cases, there were no patients in the group who never smoked, while 7 of 9 control patients had never smoked. As shown in Table 6, the lung and bronchial concentrations of chromium and nickel clearly increased with higher levels of smoking, for nickel this was especially evident in bronchial tissue. Former smokers also had elevated contents of both chromium and nickel, but to a somewhat lesser extent. The lymph node concentration of chromium was found to decrease with heavy smoking. There was no clear pattern for the nickel concentration. Among cancer cases with known smoking habits, females tended to smoke slightly more than males. Each patient was given a figure corresponding to the various smoking groups in Table 6, and mean values for males and females were calculated. The mean was 3.6 among females, compared to 2.9 in males.
Discussion Our present results indicate that the lung tissue concentration of chromium is increased in patients with bronchial cancer, independent of their former occupational history. However, only a two-fold increase was found, compared with an almost lOO-fold difference observed in deceased lung-cancer patients with a long-lasting occupational exposure to chromium 9, 10,21. Our findings are in line with a recent report from Finland2• The increased concentration of chromium is probably a reflection of a higher tobacco consumption in the cancer cases, evidenced by a possible dose relationship. Trace amounts of chromium and nickel are furthermore known to be present in tobacco smoke? The causal relationship between smoking and lung cancer development is well documented. The role of chromium and nickel in this respect is more difficult to evaluate. However, a causal relationship, either alone or acting synergistically with other carcinogens, cannot be ruled out. The effect of lowdose chromium exposure in lung carcinogenesis has earlier been hypothesized on the basis of a linear relationship between the level of hexavalent chromium exposure and cancer risk4 . Interestingly, increasing levels of tobacco consumption were found to be associated with increased concentration of chromium in broncho-pulmonary tissue, in contrast to decreasing levels in hilar lymph nodes (Table 6, groups 3, 4 and 5). This finding focuses the question of basic mechanisms. The chromium content of tobacco smoke may itself contribute to increased deposition in bronchial and lung tissue with higher levels of smoking. Secondly, it is well known that bronchial clearance mechanisms are impaired by smoking. Thirdly, it has also been documented that tobacco smoke exerts a depressive effect on the lymphatic clearance of the lung8• This may explain our findings of low chromium concentration in hilar lymph nodes and high contents in broncho-pulmonary tissue in cases with high tobacco consumption (Table 6, groups 4 and 5). Furthermore, smoking is associated with the development
Table 6. Median concentration of chromium and nickel (ng/g dry weight) with relation to different levels of smoking among cancer patients Smoking] Chromium group 0 group 1 group 2 group 3 group 4 group 5 Nickel group 0 group 1 group 2 group 3 group 4 group 5
N
Lung tissue Bronchial tissue
3 0
3441
282
9823
3 5 2
7
3382 2462 3631 3860
503 1294 1480 8668
13196 21500 16950 14999
3 0
140
62
392
156 236 556 212
401 685 1511 2196
1202 958 3728 1056
7
3 5 2
Lymph nodes
Smoking groups: 0 = no information, 1 = never smoked, 2 = former smoker, 3 = < 50 grams/week, 4 = 50-100 grams/week, 5 = > 100 grams/week. 1
of emphysema, in which chromium retention has been demonstrated 2 • These mechanisms may all playa role in the pathogenesis of chromium deposition in lung tissue. A detailed occupational history was obtained from the National Health Service, but only four patients had a possible or probable exposure to chromium. The lung tissue concentration in these was only slightly above the median value in cancer patients, thus supporting our assumption that no heavy exposure had been present. Alternatively, chromium may have been cleared from the lung tissue after the exposure had ceased. It may be argued that the increased chromium content may be caused by pathophysiological changes in the lung tissue following cancer development. A relationship with survival length should thus be expected, but this was not found. It is therefore unlikely that chromium retention should represent secondary effects of tumor development, a conclusion also reached by others 2. The concentration of nickel was significantly increased in bronchial tissue in cancer patients (about three-fold). Our findings indicate a relationship to smoking. Data on nickel deposition in organs from patients with lung cancer is at present only available in occupationally exposed individuals 1,18. Our findings demonstrate a retention of nickel in the airways, independent of occupational exposure. As for chromium, this is probably related to tobacco smoking. Females compared with males had a higher concentration of chromium and nickel in bronchial tissue and hilar lymph nodes, while the content in lung tissue was somewhat lower. The sex difference was present in both controls and cancer patients. The reason for this is not clear. Females tended to smoke somewhat more than their male counterparts, thus other factors than smoking may be involved.
722 . L. A. Akslen et al.
In conclusion, increased levels of both chromium and nickel were found in bronchial cancer patients, related to smoking and unrelated to occupational history. The relationship between high dose exposure of certain chromium and nickel compounds and lung cancer development is well established. However, a carcinogenic effect of small amounts of these metals may be present and should be further examined.
Acknowledgement We thank the Ministry for Research and Technology of the Federal Republic of Germany for financial support in the context of a research project (No. 01 HK 694). The authors are grateful to Professor Albrecht Reith for facilitating the collaboration between the University of Erlangen and the University of Bergen. The skilled technical assistance of Anne-Marie Larsen is greatly appreciated.
References I Andersen I, Svenes KB (1989) Determination of nickel in lung specimens of thirty-nine autopsied nickel workers. Int Arch Occup Environ Health 61: 289-295 2 Anttila S et al (1989) High concentrations of chromium in lung tissue from lung cancer patients. Cancer 63: 467-473 3 Bidstrup PL, Case RAM (1956) Carcinoma of the lung in workmen in bichromates-producing industry in Great Britain. Br J Indust Med 13: 260-264 4 Chovil A, Sutherland RB, Halliday M (1981) Respiratory cancer in a cohort of nickel sinter plant workers. Br J Indust Med 38:327-333 5 Enterline PE (1974) Respiratory cancer among chromate workers. J Occup Med 16: 523-526 6 Frentzel-Beyme R (1983) Lung cancer mortality of workers employed in chromate pigment factories. J Cancer Res Clin Oncol 105: 183-188 7 IARC (1986a) Tobacco smoking. In: IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 38, Lyon: IARC, 114-116
8 IARC (1986b) Tobacco smoking. In: IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 38, Lyon, IARC, pp. 183-185 9 Kim S, Iwai Y, Fujino M, Furumoto M, Sumino K, Miyasaki K (1985) Chromium-induced pulmonary cancer. Report of a case and a review of the literature. Acta Pathol Jpn 35: 643-654 10 Kishi R, Tarumi T, Uchino E, Miyake H (1987) Chromium content of organs of chromate workers with lung cancer. Am J Indust Med 11: 67-74 11 Kollmeier H, Seemann JW, Muller KM, Rothe G, Wittig P, Schejbal VB (1987) Increased chromium and nickel content in lung tissue and bronchial carcinoma. Am J Ind Med 11: 659-669 12 Langard S, Andersen A, Gylseth B (1980) Incidence of cancer among ferrochromium and ferrosilicon workers. Br J Indust Med 37: 114-120 13 Langard S, Vigander T (1983) Occurrence of lung cancer in workers producing chromium pigments. Br J Indust Med 40: 71-74 14 Magnus K, Andersen A, H0getvedt AC (1982) Cancer of respiratory organs among workers at a nickel refinery in Norway. Second report. Int J Cancer 30: 681-685 IS Pedersen E, H0getvedt AC, Andersen A (1973) Cancer of respiratory organs among workers at a nickel refinery in Norway. Int J Cancer 12: 32-41 16 Petrilli FL, de Flora S (1982) Interpretations on chromium mutagenicity and carcinogenicity. Prog Clin Bioi Res 109: 453-464 17 Raithel HJ, Ebner G, Schaller KH, Schellmann B, Valentin H (1987) Problems in establishing norm values for nickel and chromium concentrations in human pulmonary tissue. Am J Indust Med 12: 55-70 18 Raithel HJ, Schaller KH, Reith A, Svenes KB, Valentin H (1988) Investigations on the quantitative determination of nickel and chromium in human lung tissue. Int Arch Occup Environ Health 60: 55-66 19 Skerfving S, Haeger-Aronsen B (1984) Yrkessjukdomar orsakade av kemiska faktorer. In: Lundberg I, Liden C, Hogstedt (Eds), Yrkesmedicin och yrkeshudsjukdomar. Arbetarskyddstyrelsen. AB Almanna F0rlaget, Helsingborg, 61-71 20 Sunderman FW Jr (1984) Recent progress in nickel carcinogenesis. Toxicol Environ Chern 8: 235-252 21 Tsuneta Y, Ohsaki Y, Kimura K, Mikami H, Abe S, Murao M (1980) Chromium content of lungs of chromate workers with lung cancer. Thorax 35: 294-297 22 WHO (1982) The World Health Organization histological typing of lung tumors. Am J Clin Pathol 77: 123-136
Received November 22, 1989 . Accepted in revised form January 26, 1990
Key words: Chromium and nickel- Bronchial carcinoma - Human carcinogenesis - Lung cancer - Atomic absorption analysis Lars A. Akslen MD, Dept. of Pathology, The Gade Institute, Haukeland Hospital, N-5021 Bergen, Norway
Increased Content of Chromium and Nickel in Lung Tissues from Patients with Bronchial Carcinoma L. A. Akslen and A. O. Myking Department of Pathology, The Gade Institute, University of Bergen, Bergen, Norway
O. M0rkve and A. Gulsvik Department of Thoracic Medicine, University of Bergen, Bergen, Norway
H. J. Raithel and K. H. Schaller Institute of Occupational and Social Medicine and Policlinic for Occupational Diseases, University of Erlangen, Erlangen, FRG
SUMMARY
Epidemiological studies have shown that occupational exposure to certain chromium and nickel compounds is followed by an increased lung cancer incidence. However, few data exist on the content ofthese metals in lung cancer patients in general. In the present study, central and peripheral lung tissue, bronchial tissue and hilar lymph nodes were collected from 20 patients with bronchial carcinoma and 21 control individuals, and the tissue concentration ofchromium and nickel was measured by use ofatomic absorption analysis. Increased levels of both metals were found in cancer patients as compared to controls. Lung tissue concentration of chromium was two-fold increased, while the bronchial wall content of nickel was three times the level in control individuals. Smokers showed a dose-related increase in the deposition of both chromium and nickel. Furthermore, in cancer patients an inverse relationship between smoking and the tissue level ofchromium in regional lymph nodes was found, possibly indicating a depressive effect on pulmonary clearance mechanisms. Our results emphasize the possible role of small amounts of chromium and nickel as agents in bronchial carcinogenesis, unrelated to occupation and probably related to tobacco smoking.
Introduction Several epidemiological studies have shown that occupational exposure to hexavalent chromium3,5,6, 12, 13 and nickel compounds4, 14, 15 is associated with increased incidence of lung cancer. In addition, very high contents of chromium have been found in lung tissues and other organs from persons with a known and long-lasting exposure9, 10, 21. Mutagenicity of chromium compounds has been well documented, and the interaction with nucleic © 1990 by Gustav Fischet Verlag, Stuttgart
acids may be the basis for its carcinogenic influence l6 • Research on nickel has also shown a wide range of toxicological and carcinogenic effects20 • Although occupational hazards of chromium and nickel are well established, information on the tissue content of these metals in cancer patients is very sparse. However, recent observations indicate that patients with lung cancer may have an increased content of chromium in lung tissue 2, 11. The role of this finding in bronchial carcinogenesis still remains unknown. The aim of the present study was to investigate 0344-0338/90/0186-0717$3.50/0
718 . L. A. Akslen et al.
the content and distribution of chromium and nickel in the lung, bronchial and tumor tissue as well as hilar lymph nodes in a group of unselected lung cancer patients, compared to a control group. Material and Methods
Patients Twenty concomitant patients with bronchial carcinoma were included during a 6 month period in 1986, of these 15 males and 5 females (median age 62 yrs). In four cases (2 males, 2 females) surgical specimens were examined, while the rest were autopsy cases. All patients had been hospitalized in the Department of Thoracic Medicine, Haukeland Hospital, Bergen. Bronchial carcinoma was the main cause of death in the autopsy cases. Altogether 21 control patients were included during the same period, 9 males and 12 females (median age 72 yrs). The main causes of death were ischaemic heart disease (n = 10), cerebrovascular disease (n = 6), chronic respiratory disease (n = 2), aortic stenosis (n = 1) and glioblastoma multiforme (n = 1) without tumor spread. A standardized dissection was performed in all cases. Lung cancer patients and controls were from a limited geographical area in and around Bergen at the west coast of Norway. No production of chromium and nickel is present in this area. A detailed occupational history was obtained for each patient, based on the hospital records and on information from the National Health Service. Occupational exposure to chromium and nickel 19 was graded as none, possible, and probable. Details on smoking habits were available in 17 of 20 cancer cases and 9 of 21 controls, and the quantity is given as grams per week. Histological type of tumor, according to the WHO classification 22 as well as patient survival (months) in fatal cases were registrated.
Tissue Sampling Lung tissue was collected from central and peripheral parts of each lobe, giving a total of 10 specimens from each autopsy case. Instruments free of chromium and nickel were used. In addition, specimens from the central bronchial walls and hilar lymph nodes from both sides were included, together with tumor tissue in cancer cases. In the surgical specimens, sampling from tissue surfaces and resection margins was avoided. Although contamination from surgical instruments cannot be totally excluded, the content of chromium and nickel in these four cases were found to be within the range of the autopsy cases. The specimens, most of which had a dry weight of 20 mg-300 mg, were fixed in absolute alcohol. The analysis of chromium and nickel (ng/g dry weight) was performed at the University of Erlangen (FRG) as described by Raithel et alY. Metal eluation and contamination during specimen storage in alcohol has been analyzed and found to be negligible. The variation coefficient of repeated measurements was 6% for chromium and 9.5% for nickel.
Statistical Methods The median values of chromium and nickel concentration in each tissue type (lung, bronchial, lymphoid, tumor) were calculated for each patient and used in further analysis of differences between cancer cases and control individuals. Differences were compared using the non-parametric Mann-Whitney rank-sum test. Association between different variables is expressed by the non-parametric correlation coefficient Spearman's rho (r), with
corresponding p-values. Probabilities less than 0.05 were considered significant.
Results
There was a marked intraindividual variation in the content qf chromium and nickel in lung tissue (these findings will be presented in a separate publication). Briefly, the mean ratio between maximum and minimum values was as follows: chromium among cancer cases 12.4 (range 2.9-43.3), chromium among controls 8.7 (range 2.3-21.3), nickel among cancer cases 14.1 (range 2.8-67.3) and nickel among controls 11.4 (range 2.6-29.7).
Chromium Concentration The concentration of chromium in lung tissue from cancer patients was more than twice that of controls (Table 1, Fig. 1, P = 0.031). The difference was present among both males (ratio cases/controls 1.5, p = 0.07) and females (ratio cases/controls 3.0, p = 0.19), but the numbers were small for statistical analysis (Table 2). No significant differences were found in bronchial or lymphoid tissues, but cancer cases tended to have somewhat higher concentrations than control individuals. Tumor tissue contained about twice as much chromium as bronchial tissue, but the concentration was clearly lower than in the lung tissue.
LUNG TISSUE
-
~
Cl C1l
.a 12000
•
~
Cl
•
Cl C
z
0
.... 0.05).
The chromium concentration was highest in lymphoid tissue, being about 4.1 times that of lung tissue in cancer cases and 6.1 times among controls. The lowest concentration was found in bronchial tissue, with a bronchiaVlung ratio of 0.19 and 0.24 in cancer cases and controls, respectively. There was no covariation between chromium concentrations of any tissue among control individuals, except a positive correlation between bronchial tissue and lymph nodes (r = 0.47, p < 0.05). In cancer cases, the bronchial tissue concentration was negatively correlated to the content in tumor tissue (r = -0.39, p < 0.05). The lung tissue concentration of chromium was clearly higher in males than females, in cancer cases as well as controls (Table 2). The male/female ratio among controls was 2.4 and 1.2 for cancer patients. However, in lymph nodes and bronchial tissue, females had the highest concentrations in both patient groups.
Table 3. Median concentration of nickel (ng/g dry weight) in various tissue types among cancer patients (n = 20) and control individuals (n = 21)
Tissue type Lung tissue cases controls
Males (N)
Females (N)
P-value l
Lung tissue cases controls
196 165
94 90
507 258
20 21
Bronchial tissue cases controls
416 142
71 36
1312 211
20 20
Lymphoid tissue cases controls
958 1084
477 261
2098 2087
19 21
191
97
395
13
NS NS
Bronchial tissue cases controls
546 (15) 189 (9)
1294 (5) 440 (12)
NS NS
Lymphoid tissue cases controls
12081 (14) 9335 (9)
22000 (5) 15639 (12)
1310 (12)
2020 (5)
Tumor tissue 1
Mann-Whitney rank-sum test (NS denotes p
0.037 NS NS
> 0.05).
p-
value l NS
Tumor tissue
0.026
NS
Ql = Lower quartile; Q3 = Upper quartile, N = Number of cases.
1
Mann-Whitney rank-sum test (NS denotes p
> 0.05).
BRONCH IAL TISSUE
-
10 4
~
01 QI
~
...>. '0
01
10
•
••
•
•
•
3
~ r
01
c:
0
i=
« a::: ~
Z
w u
102
It
I
z
0
2851 (5) 957 (12)
N
value
U
3441 (15) 2253 (9)
Qr
value
z
Table 2. Median concentration of chromium (ng/g dry weight) in various tissue types and specified for both sexes (N = number of cases)
Ql
Median value
Tissue type
I
~ • •
•
I
...J
W
:::.::
u Z
10 1
1
• i
CASES
i
CONTROLS
Fig. 2. Individual log-values of nickel concentration in bronchial tissue from cancer cases and control patients (median values indicated).
720 . L. A. Akslen et al.
tion of nickel was slightly higher in cancer cases, but the difference was not statistically significant (Table 3). However, in lymphoid tissue there was no difference between cancer cases and controls. Tumor tissue showed values comparable to that of lung tissue. Similar to chromium, the nickel concentration was highest in lymphoid tissue, 6.6 times that of lung tissue in control patients and 4.9 among cancer cases. In contrast to chromium, the deposition of nickel in the bronchial walls was about twice that in lung parenchyma among cancer cases, while there was no such difference among controls. The nickel concentration in various tissue types was significantly correlated, in both cancer patients and controls. The strongest correlation in control individuals was lung - bronchial tissue (r = 0.67, p < 0.005) and lung lymphoid tissue (r = 0.67, p < 0.001). In cancer cases, the lung - bronchial correlation (r = 0.78, p < 0.001) and bronchial tissue - lymph node correlation (r = 0.66, p < 0.005) were strong as well. The male cancer cases tended to have a higher lung tissue concentration of nickel than females, without such sex difference among controls (Table 4). In bronchial and lymphoid tissue females had somewhat higher values, in both cancer patients and control individuals, without being statistically significant.
Chromium and Nickel A significant correlation between the tissue concentration of chromium and nickel was found in lung (r = 0.51, p < 0.05) and bronchial tissue (r = 0.43, P < 0.05) among cancer cases, and in bronchial tissue (r = 0.46, p < 0.05) and lymph nodes (r = 0.79, p < 0.001) among control individuals.
Age The median age of cancer patients was 62 yrs, as compared to 72 yrs for control patients (p = 0.0014). Among cancer patients, the median age of males and females were 64 yrs and 58 yrs, respectively. The corresponding figures for control individuals were 72 yrs and 78 yrs. Within each group the age differences were not statistically significant. There was no significant correlation between age and concentration of chromium and nickel in any tissue, neither in cancer patients nor in controls.
Histology There was some variation between different histological tumor types. However, the number of cases in each group was small. The lung tissue concentration of chromium was rather low in the three cases of small cell carcinoma, with a median value of 2279 (ng/g dry weight), compared to 3129 (ng/g) in adenocarcinomas (n = 6) and 3898 (ng/g) in squamous cell carcinomas (n = 8). The nickel concentration in bronchial tissue was high in unclassified carcinoma (2464 ng/g; n = 3), compared to squamous cell carcinoma (510 ng/g) , adenocarcinoma (291 ng/g) and small cell carcinoma (401 ng/g).
Table 4. Median concentration of nickel (ng/g dry weight) in various tissue types and specified for both sexes (N = number of cases Males (N)
Tissue type Lung tissue cases controls Bronchial tissue cases controls
Females (N)
P-value 1
243 (15) 165 (9)
142 (5) 154 (12)
NS NS
401 (15) 71 (9)
431 (5) 159 (11)
NS NS
Lymphoid tissue cases controls
882 (14) 478 (9)
1075 (5) 1377 (12)
NS NS
Tumor tissue
191 (9)
227 (4)
NS
1
Mann-Whitney rank-sum test (NS denotes p > 0.05).
Survival The median survival from the time of cancer diagnosis in all fatal cases (n = 16) was 6 months (range 1-24 months). The concentration of chromium in bronchial tissue was positively correlated to increasing survival time (r = 0.72, p < 0.005), while the concentration of nickel in lung tissue increased somewhat with survival (r = 0.50, p < 0.05). No other statistically significant correlations were found.
Occupational History As judged by a detailed occupational history, four males, three of them having cancer, had a possible or probable exposure to chromium, and one of these cases (no. 15) also had a possible exposure to nickel (Table 5). The lung tissue concentration of chromium among the three cancer cases was 1.6 to 4.0 times that of controls, and the mean value was 1.3 times that of other cancer patients. The one control individual had a chromium concentration of 1.8 times that of other controls. The one cancer patient with a possible exposure to nickel had low tissue concentrations. Table 5. Concentrations of chromium and nickel (ng/g dry weight) in four patients with possible or probable occupational exposure Tissue concentration Patients
Group Exposure l Lung tissue
Bronchial Lymphoid tissue tissue
Chromium Case 2 Case 15 Case 26 Case 41
CA CA CA CO
(+) (+)
6734 2736 3476 3073
1480 282 165 556
21080 9823 13191 15609
Nickel Case 15
CA
(+)
140
62
198
CA = cancer patient, CO = control patient. possible, + = probable.
1
Exposure: (+) =
Chromium and Nickel in Lung Cancer' 721
Smoking Habits Among cancer cases, there were no patients in the group who never smoked, while 7 of 9 control patients had never smoked. As shown in Table 6, the lung and bronchial concentrations of chromium and nickel clearly increased with higher levels of smoking, for nickel this was especially evident in bronchial tissue. Former smokers also had elevated contents of both chromium and nickel, but to a somewhat lesser extent. The lymph node concentration of chromium was found to decrease with heavy smoking. There was no clear pattern for the nickel concentration. Among cancer cases with known smoking habits, females tended to smoke slightly more than males. Each patient was given a figure corresponding to the various smoking groups in Table 6, and mean values for males and females were calculated. The mean was 3.6 among females, compared to 2.9 in males.
Discussion Our present results indicate that the lung tissue concentration of chromium is increased in patients with bronchial cancer, independent of their former occupational history. However, only a two-fold increase was found, compared with an almost lOO-fold difference observed in deceased lung-cancer patients with a long-lasting occupational exposure to chromium 9, 10,21. Our findings are in line with a recent report from Finland2• The increased concentration of chromium is probably a reflection of a higher tobacco consumption in the cancer cases, evidenced by a possible dose relationship. Trace amounts of chromium and nickel are furthermore known to be present in tobacco smoke? The causal relationship between smoking and lung cancer development is well documented. The role of chromium and nickel in this respect is more difficult to evaluate. However, a causal relationship, either alone or acting synergistically with other carcinogens, cannot be ruled out. The effect of lowdose chromium exposure in lung carcinogenesis has earlier been hypothesized on the basis of a linear relationship between the level of hexavalent chromium exposure and cancer risk4 . Interestingly, increasing levels of tobacco consumption were found to be associated with increased concentration of chromium in broncho-pulmonary tissue, in contrast to decreasing levels in hilar lymph nodes (Table 6, groups 3, 4 and 5). This finding focuses the question of basic mechanisms. The chromium content of tobacco smoke may itself contribute to increased deposition in bronchial and lung tissue with higher levels of smoking. Secondly, it is well known that bronchial clearance mechanisms are impaired by smoking. Thirdly, it has also been documented that tobacco smoke exerts a depressive effect on the lymphatic clearance of the lung8• This may explain our findings of low chromium concentration in hilar lymph nodes and high contents in broncho-pulmonary tissue in cases with high tobacco consumption (Table 6, groups 4 and 5). Furthermore, smoking is associated with the development
Table 6. Median concentration of chromium and nickel (ng/g dry weight) with relation to different levels of smoking among cancer patients Smoking] Chromium group 0 group 1 group 2 group 3 group 4 group 5 Nickel group 0 group 1 group 2 group 3 group 4 group 5
N
Lung tissue Bronchial tissue
3 0
3441
282
9823
3 5 2
7
3382 2462 3631 3860
503 1294 1480 8668
13196 21500 16950 14999
3 0
140
62
392
156 236 556 212
401 685 1511 2196
1202 958 3728 1056
7
3 5 2
Lymph nodes
Smoking groups: 0 = no information, 1 = never smoked, 2 = former smoker, 3 = < 50 grams/week, 4 = 50-100 grams/week, 5 = > 100 grams/week. 1
of emphysema, in which chromium retention has been demonstrated 2 • These mechanisms may all playa role in the pathogenesis of chromium deposition in lung tissue. A detailed occupational history was obtained from the National Health Service, but only four patients had a possible or probable exposure to chromium. The lung tissue concentration in these was only slightly above the median value in cancer patients, thus supporting our assumption that no heavy exposure had been present. Alternatively, chromium may have been cleared from the lung tissue after the exposure had ceased. It may be argued that the increased chromium content may be caused by pathophysiological changes in the lung tissue following cancer development. A relationship with survival length should thus be expected, but this was not found. It is therefore unlikely that chromium retention should represent secondary effects of tumor development, a conclusion also reached by others 2. The concentration of nickel was significantly increased in bronchial tissue in cancer patients (about three-fold). Our findings indicate a relationship to smoking. Data on nickel deposition in organs from patients with lung cancer is at present only available in occupationally exposed individuals 1,18. Our findings demonstrate a retention of nickel in the airways, independent of occupational exposure. As for chromium, this is probably related to tobacco smoking. Females compared with males had a higher concentration of chromium and nickel in bronchial tissue and hilar lymph nodes, while the content in lung tissue was somewhat lower. The sex difference was present in both controls and cancer patients. The reason for this is not clear. Females tended to smoke somewhat more than their male counterparts, thus other factors than smoking may be involved.
722 . L. A. Akslen et al.
In conclusion, increased levels of both chromium and nickel were found in bronchial cancer patients, related to smoking and unrelated to occupational history. The relationship between high dose exposure of certain chromium and nickel compounds and lung cancer development is well established. However, a carcinogenic effect of small amounts of these metals may be present and should be further examined.
Acknowledgement We thank the Ministry for Research and Technology of the Federal Republic of Germany for financial support in the context of a research project (No. 01 HK 694). The authors are grateful to Professor Albrecht Reith for facilitating the collaboration between the University of Erlangen and the University of Bergen. The skilled technical assistance of Anne-Marie Larsen is greatly appreciated.
References I Andersen I, Svenes KB (1989) Determination of nickel in lung specimens of thirty-nine autopsied nickel workers. Int Arch Occup Environ Health 61: 289-295 2 Anttila S et al (1989) High concentrations of chromium in lung tissue from lung cancer patients. Cancer 63: 467-473 3 Bidstrup PL, Case RAM (1956) Carcinoma of the lung in workmen in bichromates-producing industry in Great Britain. Br J Indust Med 13: 260-264 4 Chovil A, Sutherland RB, Halliday M (1981) Respiratory cancer in a cohort of nickel sinter plant workers. Br J Indust Med 38:327-333 5 Enterline PE (1974) Respiratory cancer among chromate workers. J Occup Med 16: 523-526 6 Frentzel-Beyme R (1983) Lung cancer mortality of workers employed in chromate pigment factories. J Cancer Res Clin Oncol 105: 183-188 7 IARC (1986a) Tobacco smoking. In: IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 38, Lyon: IARC, 114-116
8 IARC (1986b) Tobacco smoking. In: IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, Vol. 38, Lyon, IARC, pp. 183-185 9 Kim S, Iwai Y, Fujino M, Furumoto M, Sumino K, Miyasaki K (1985) Chromium-induced pulmonary cancer. Report of a case and a review of the literature. Acta Pathol Jpn 35: 643-654 10 Kishi R, Tarumi T, Uchino E, Miyake H (1987) Chromium content of organs of chromate workers with lung cancer. Am J Indust Med 11: 67-74 11 Kollmeier H, Seemann JW, Muller KM, Rothe G, Wittig P, Schejbal VB (1987) Increased chromium and nickel content in lung tissue and bronchial carcinoma. Am J Ind Med 11: 659-669 12 Langard S, Andersen A, Gylseth B (1980) Incidence of cancer among ferrochromium and ferrosilicon workers. Br J Indust Med 37: 114-120 13 Langard S, Vigander T (1983) Occurrence of lung cancer in workers producing chromium pigments. Br J Indust Med 40: 71-74 14 Magnus K, Andersen A, H0getvedt AC (1982) Cancer of respiratory organs among workers at a nickel refinery in Norway. Second report. Int J Cancer 30: 681-685 IS Pedersen E, H0getvedt AC, Andersen A (1973) Cancer of respiratory organs among workers at a nickel refinery in Norway. Int J Cancer 12: 32-41 16 Petrilli FL, de Flora S (1982) Interpretations on chromium mutagenicity and carcinogenicity. Prog Clin Bioi Res 109: 453-464 17 Raithel HJ, Ebner G, Schaller KH, Schellmann B, Valentin H (1987) Problems in establishing norm values for nickel and chromium concentrations in human pulmonary tissue. Am J Indust Med 12: 55-70 18 Raithel HJ, Schaller KH, Reith A, Svenes KB, Valentin H (1988) Investigations on the quantitative determination of nickel and chromium in human lung tissue. Int Arch Occup Environ Health 60: 55-66 19 Skerfving S, Haeger-Aronsen B (1984) Yrkessjukdomar orsakade av kemiska faktorer. In: Lundberg I, Liden C, Hogstedt (Eds), Yrkesmedicin och yrkeshudsjukdomar. Arbetarskyddstyrelsen. AB Almanna F0rlaget, Helsingborg, 61-71 20 Sunderman FW Jr (1984) Recent progress in nickel carcinogenesis. Toxicol Environ Chern 8: 235-252 21 Tsuneta Y, Ohsaki Y, Kimura K, Mikami H, Abe S, Murao M (1980) Chromium content of lungs of chromate workers with lung cancer. Thorax 35: 294-297 22 WHO (1982) The World Health Organization histological typing of lung tumors. Am J Clin Pathol 77: 123-136
Received November 22, 1989 . Accepted in revised form January 26, 1990
Key words: Chromium and nickel- Bronchial carcinoma - Human carcinogenesis - Lung cancer - Atomic absorption analysis Lars A. Akslen MD, Dept. of Pathology, The Gade Institute, Haukeland Hospital, N-5021 Bergen, Norway
