Subsequent development of cholangiocarcinoma caused by exposure to 1,2-dichloropropane and/or dichloromethane in the printing company in Osaka, Japan

Abstract

After the report of 17 patients with occupational cholangiocarcinoma caused by long-term exposure to high concentrations of 1,2-dichloropropane and/or dichloromethane in a printing company in Osaka in 2014, additional five patients were diagnosed to have such cholangiocarcinoma. Cholangiocarcinoma was detected during regular health examination or follow-up for liver dysfunction in four of the five patients. Nearly all five patients presented with clinicopathological findings such as an elevated γ-glutamyl transpeptidase activity at the diagnosis, regional dilatation of intrahepatic bile ducts without tumor-induced obstruction, chronic bile duct injury, and precancerous/early cancerous lesions (biliary intraepithelial neoplasia and intraductal papillary neoplasm of the bile duct) at various sites of the bile duct. These findings were similar to those of the previous 17 patients. In total, cholangiocarcinoma developed in 22 of 95 workers exposed to 1,2-dichloropropane in the printing company. Of 22 patients with cholangiocarcinoma, 18 patients were members of 19 high exposure workers (≥1,500 ppm-yr). These findings strengthen further the theory that 1,2-dichloropropane causes occupational cholangiocarcinoma. Regular health examination of workers exposed to 1,2-dichloropropane and/or dichloromethane is necessary to detect such cholangiocarcinoma because the potential of the carcinogenesis risk persists over the long term.

Introduction

An outbreak of cholangiocarcinoma at an offset color-proof printing department in a printing company in Osaka was reported in 2013, and the clinicopathological findings of these patients with cholangiocarcinoma were reported in 2014^{1, 2)}. Such cholangiocarcinoma suspected to be caused by long-term exposure to high concentrations of 1,2-dichloropropane (DCP) and/or dichloromethane (DCM) was recognized as an occupational disease (occupational cholangiocarcinoma) by the Japanese Ministry of Health, Labour and Welfare in 2013³⁾. Thus far, the ministry has identified approximately 60 patients with occupational cholangiocarcinoma all over Japan. In June 2014, the International Agency for Research on Cancer decided to classify DCP as group 1 (carcinogenic to humans) and DCM as group 2A (probably carcinogenic to humans)⁴⁾. Occupational exposure to DCP and/or DCM was added as a risk factor for intrahepatic cholangiocarcinoma in the WHO classification of tumors in 2019⁵⁾.

In the previous report²⁾, 17 patients were diagnosed with cholangiocarcinoma between November 1996 and November 2012. The characteristics of occupational cholangiocarcinoma caused by the long-term exposure to high concentrations of DCP and/or DCM were as follows: (1) elevated activities of γ-glutamyl transpeptidase (γ-GTP), aspartate aminotransferase (AST), and alanine aminotransferase (ALT) at the time of the cholangiocarcinoma diagnosis and elevated activities in some patients several years prior to the diagnosis; (2) regional dilatation of intrahepatic bile ducts without tumor-induced obstruction in diagnostic imaging; (3) chronic bile duct injury with DNA damage, precancerous/early cancerous lesions of the bile ducts such as biliary intraepithelial neoplasia (BilIN) and intraductal papillary neoplasm of the bile duct (IPNB) in various sites of the large bile ducts, and invasive cholangiocarcinoma such as mass-forming type and invasive IPNB and mainly papillary-type extrahepatic cholangiocarcinoma on the pathological examination; (4) hypermutation, substantial strand bias, and unique trinucleotide mutational changes in the whole-exome analysis of the cancerous tissues. The multicentric carcinogenesis including recurrence from multicentric origin is also characteristic because of high carcinogenic potential caused by DNA-injured bile ducts^{2, 6,7,8,9,10,11,12,13,14,15,16,17)}.

In the printing company, DCP was used from 1987 until 2006 and DCM was used from 1987 until 1996. In an epidemiological study, the standardized incidence ratio of cholangiocarcinoma in 95 workers exposed to DCP was 1,171 (95% CI, 682–1,875), based on the observation from 1987 to 2012, and an exposure–response relationship was found between cumulative exposure to DCP (ppm-yr) and cholangiocarcinoma¹⁸⁾.

Because the frequency of DCP use with high occupational exposure in the printing industry in many countries, except Japan, Korea, and middle east, is low¹⁹⁾, confirming that DCP is responsible for occupational cholangiocarcinoma is challenging in many countries, such as United States and European countries. In addition, DCP has not been used in the printing industry in Japan since 2013, thus, it is unlikely that similar cases will occur in the future. Therefore, information about the subsequent development of occupational cholangiocarcinoma in the printing company in Osaka after the previous report²⁾ is important to reconfirm that DCP and/or DCP cause occupational cholangiocarcinoma.

After the previous report²⁾, cholangiocarcinoma was newly detected in five patients who were exposed to DCP and/or DCM at the same printing company at 2015, 2017, 2018, 2022, and 2023. In total, cholangiocarcinoma developed in 22 workers by the end of 2023. In this report, the clinicopathological findings and exposure condition to DCP in the five patients with reference to those of the patients in previous reports^{2, 18)} are described.

Patients and Methods

The five patients reported herein were former or current workers at the offset-color proof-printing department at the same printing company, and they were treated at three hospitals in Osaka. Clinical findings, chemical exposure levels, laboratory test results, diagnostic imaging, pathological findings, treatments, and prognosis of the five patients were investigated. The clinicopathological findings in two patients (patients 2 and 4) has been reported previously^{14, 17)}. Although one patient (patient 2) did not apply as to have occupational cholangiocarcinoma that is certified by the Ministry of Health, Labour and Welfare in Japan owing to his will and lack of a legally authorized representative, the patient assumed to have occupational cholangiocarcinoma owing to his history of exposure to high concentrations of DCP and DCM while working for 6 yr 1 month at the printing company, in addition to typical clinicopathological findings of occupational cholangiocarcinoma. Since 2013, regular health examination (twice a year) comprising measurement of total bilirubin, AST, ALT, γ-GTP, and carbohydrate 19-9 (CA19-9) and imaging diagnosis, including abdominal ultrasonography and computed tomography, for workers exposed to DCP for at least 2 yr are being conducted by the Ministry of Health, Labour and Welfare (regular health examination due to the high risk for occupational cholangiocarcinoma using the health handbook).

The pathological findings were recorded and described according to the World Health Organization’s classifications for intrahepatic and extrahepatic cholangiocarcinoma⁵⁾. Precancerous/early cancerous lesions of the bile ducts were classified as BilIN and IPNB. Chronic bile duct injury was used as a collective lesion of various injuries such as epithelial damages, duct wall and periductal tissue fibrosis, focal bile duct loss, biliary epithelial hyperplasia, and chronic inflammation cell infiltration²⁾. Immunological staining was performed using primary antibodies against γ-H2AX (1:100 Rabbit Monoclonal; Novus Biologicals, Littleton, Co., Centennial, CO, USA), which is a marker for double-strand DNA injuries, and S100P (1:100 Rabbit Monoclonal, Abcam, Tokyo, Japan), which is a marker for malignant transformation and large duct type (subclassification of intrahepatic cholangiocarcinoma). Cholangiocarcinoma was staged according to the TNM classification of malignant tumours (eighth edition) by UICC International Union against Cancer²⁰⁾.

The cumulative exposure to DCP and DCM was calculated in 95 workers (78 men and 17 women), including 17 patients in the previous report²⁾ and five patients in this report, at the color-proof printing department of the printing company, according to the methods reported previously¹⁸⁾. In brief, monthly exposure concentrations of DCP and DCM from 1987 to 2006 were estimated using the results of reproducibility experiments and monthly actual usage of these chemicals in the department, and by combining the estimated concentrations with the duration of employment for each worker, cumulative exposure was calculated (for more details, see reference¹⁸⁾).

Of 95 present or previous workers at the printing company, we could follow-up 45 workers (47.4%), including 22 workers with occupational cholangiocarcinoma, at the end of 2023.

This study was conducted according to the guidelines of the ethics committee of Osaka Metropolitan University and the Declaration of Helsinki and was approved by the ethics committee of Osaka Metropolitan University (No. 2022-116, 2023-102). All of the subjects provided written informed consent, including presentation as case reports.

Results

Clinical findings

The age of the five patients at cholangiocarcinoma diagnosis ranged from 41 to 52 yr (mean, 46 yr, Table 1). Of the five patients, four patients were male and one was female. Patients 1–4 were exposed to DCP (exposure duration, 6 yr 1 month to 15 yr 2 months; cumulative exposure, 1,503 to 3,433 ppm-yr) and DCM (exposure duration, 1 yr 2 months to 5 yr 8 months; cumulative exposure, 315 to 1,109 ppm-yr). Patient 5 was exposed to DCP (exposure duration, 8 yr 4 months; cumulative exposure, 1,952 ppm-yr). Patient 2 visited the hospital because of jaundice and appetite loss and did not undergo a regular health examination. Of the other four patients, cholangiocarcinoma was detected during follow-up and treatments for liver dysfunction (high serum γ-GTP activity) in patients 1, 3, and 4. In one patient (patient 5), cholangiocarcinoma was detected during the regular health examination according to the health handbook by the Ministry of Health, Labour and Welfare. Patient 4 was followed and treated for liver dysfunction in Sakai City Medical Center and underwent regular health checkups arranged by the company. Patient 2 was a habitual alcohol consumer (≥80 g of ethanol daily), and patients 1 and 3 were smokers, with Brinkman index was 810 and 400, respectively.

Table 1. Clinical findings in patients with occupational cholangiocarcinoma.

Patient no.	Time of diagnosis (yr)	Age/sex	Exposed organic solvents	Exposure duration (yr)		Cummulative exposure (ppm-yr)		Latency periods (yr)	Symptom or health examination	Alcohol abuse	Smoking	Body mass index
				DCP	DCM	DCP	DCM
1	2015	48/M	DCP, DCM	13.2	5.7	3,083	1,109	24.7	Liver dysfunction	No	Yes	31.5
2	2017	41/M	DCP, DCM	6.1	2.8	1,503	760	24.1	Jaundice, appetite loss	Yes	No	23.2
3	2018	47/M	DCP, DCM	7.3	1.2	1,898	315	23.6	High serum concentration of CA19-9	No	Yes	22.2
4	2022	52/M	DCP, DCM	15.2	4.6	3,433	1,020	30.6	Liver tumor	No	No	25.5
5	2023	44/F	DCP	8.3		1,952		25.7	Liver dysfunction, high serum concentration of CA19-9	No	No	30.4

DCP: 1,2-dichloropropane; DCM: dichloromethane; CA19-9: carbohydrate antigen 19-9.

Laboratory test results

The laboratory test results at the diagnosis of cholangiocarcinoma are shown in Table 2. The serum concentrations of total bilirubin increased in patient 2 with obstructive jaundice. Serum activities of AST and/or ALT were high in patients 1–3. The serum γ-GTP activity was also high in all patients. In patients 3–5, for whom laboratory test results from several years before the diagnosis of cholangiocarcinoma, the serum γ-GTP activities were consistently high or gradually increased, followed by an increase in AST and/or ALT activities (Supplementary Fig. 1). At the diagnosis of cholangiocarcinoma, three patients had high concentration of CA19-9. The serum concentration of carcinoembryonic antigen was high in patients 2 and 5. The tests for serum hepatitis B surface antigen and anti-hepatitis C antibody were negative in all patients.

Table 2. Laboratory test results and diagnosis of cholangiocarcinoma.

Patient no.	Laboratory tests								Diagnostic imaging	Diagnosis of cholangiocarcinoma
	Total bilirubin (mg/dL)(0.4–1.5)1	AST (IU/L)(13–30)1	ALT (IU/L)(7–23)1	γ-GTP (IU/L)(9–32)1	CEA (ng/mL)(≤5)1	CA19-9 (U/mL)(≤37)	HBsAg	HCVAb	Methods for clinical diagnosis	Location and type of cholangiocarcinoma (main tumor)	Stage by TNM classification20)
1	0.4	18	37	77	2.6	15	Negative	Negative	CT	Extrahepatic, flat-infiltrating type	IIB
2	10.7	76	226	319	17.9	2	Negative	Negative	CT	Extrahepatic, uncertain2Intrahepatic, periductal infiltration type	IV
3	0.8	59	103	600	3	889	Negative	Negative	CT	Intrahepatic, mass-forming type	IIIB
4	0.74	19	23	93	1.6	36.1	Negative	Negative	CT, MRI	Intrahepatic, mass-forming type	II
5	0.2	24	23	115	5.1	1,182	Negative	Negative	MRI	Intrahepatic, mass-forming type	II

¹The reference range.

²The type of cholangiocarcinoma was uncertain because a bulky mass at the hepatoduodenal ligament involved the main tumor and lymph node metastasis, spreading the common duct and invading the duodenum, pancreas, and liver.

Diagnosis and pathological findings

The results of diagnostic imaging are shown in Table 2 and Fig. 1. The thickness of the extrahepatic bile duct wall on computed tomography (CT) and duct stenosis on magnetic resonance (MR) cholangiopancreatography were detected in patient 1. The intrahepatic mass lesions were detected on CT and/or MR imaging (MRI) in patients 3–5. In patient 2, CT exhibited dilatation of the intrahepatic bile ducts with the common bile duct obstruction owing to a bulky mass in the hepatoduodenal ligament as well as enlarged para-aortic lymph nodes. Although the intrahepatic bile ducts were entirely dilated, cystic dilatation of the proximal side of the biliary branch in segment 2 with the intraductal tumor suspected as IPNB was identified on CT and MR cholangiopancreatography. The lymph nodes swelling was detected in patients 1–3. The regional dilatation of intrahepatic bile ducts without tumor-induced obstruction was found in patients 3–5. Such diagnostic imaging was useful to distinguish cholangiocarcinoma from other hepatobiliary diseases. Fluorodeoxyglucose-positron emission tomography (FDG-PET) was performed in patients 1, 3, 4, and 5 and positive accumulation was demonstrated in the main tumors in patients 1, 3, and 5 but not in patient 4. The positive accumulation in the lymph nodes was demonstrated in patients 1 and 3. Precancerous/early cancerous lesions were not demonstrated on FDG-PET.

Fig. 1.

Diagnostic imaging of patients with occupational cholangiocarcinoma.

(a) The stenosis (arrows) of the extrahepatic bile duct on magnetic resonance cholangiopancreatography in patient 1; (b) intrahepatic cholangiocarcinoma of the mass-forming type (arrow) in patient 3; (c) lymph node metastasis (arrow) in patient 3; (d) positive accumulation in the main tumor (arrow) on Fluorodeoxyglucose-positron emission tomography (FDG-PET) in patient 3; (e) positive accumulation in the lymph node (arrow) on FDG-PET in patient 3; (f) regional dilated intrahepatic bile duct without tumor-induced obstruction (arrows) in patient 3; (g) bulky mass (arrow) around the hepatoduodenal ligament with involvement of the common bile duct and the pancreas head in patient 2 [14]; (h) cystic dilatation of the proximal side of the bile duct in segment 2 with the intraductal tumor (arrow) suspected as intraductal papillary neoplasm of the bile duct in patient 2 [14]; (i) intrahepatic cholangiocarcinoma of mass-forming type (arrow) in patient 5.

Pathological findings

Pathological findings are shown in Tables 2 and 3. Of the five patients, patients 3–5 had intrahepatic cholangiocarcinoma, patient 1 had extrahepatic cholangiocarcinoma, and patient 2 had both intra- and extrahepatic cholangiocarcinoma. Of the four patients with intrahepatic cholangiocarcinoma, the tumors were classified as a mass-forming type in patients 3–5 and periductal infiltration type in patient 2. Of the two patients with extrahepatic cholangiocarcinoma, patient 1 was classified as having a flat-infiltration type, whereas in patient 2, it could not be classified because the bulky mass in the hepatoduodenal ligament involved the main tumor and lymph node metastasis, spreading to the common bile duct and invading the duodenum, pancreas, and liver.

Table 3. Pathological findings of occupational cholangiocarcinoma.

Patient no.	Main tumor	Lymph node metastasis	Chronic bile duct injury	BilIN	IPNB
1	Moderately differentiated adenocarcinoma	Yes	Yes	NE	NE
2	Extrahepatic cholangiocarcinoma: Poorly differentiated adenocarcinoma with partial squamous cell differentiationIntrahepatic cholangiocarcinoma: well-differentiated adenocarcinoma	Yes	Yes	Yes	Yes
3	Poorly differentiated adenocarcinoma	Yes	Yes	Yes	Yes
4	Well-differentiated adenocarcioma	No	Yes	Yes	No
5	Moderately differentiated adenocarcinoma	No	Yes	Yes	Yes

BilIN: biliary intraepithelial neoplasia; IPNB: intraductal papillary neoplasm of the bile duct; NE: BilIN and IPNB could not be evaluated because of the absence of nearly all the biliary epithelium.

As a representative case, the pathological findings in patient 5 are shown in Fig. 2. The mass-forming type of intrahepatic cholangiocarcinoma (patients 3–5) showed well- and poorly differentiated cholangiocarcinoma. Patient 1 with extrahepatic cholangiocarcinoma had moderately differentiated adenocarcinoma, and patient 2 had poorly differentiated adenocarcinoma with partial squamous cell differentiation. The lymph node metastases in the hepatoduodenal ligament, peripancreatic lesion, and/or along the common hepatic artery were found in patients 1–3.

Fig. 2.

Pathological findings of occupational cholangiocarcinoma in patient 5.

(a) Intrahepatic cholangiocarcinoma of mass-forming type, operative specimen; (b) moderately differentiated adenocarcinoma, Hematoxylin and eosin (H&E) staining, 200× magnification; (c) mucin production of the tumor, Alcian blue staining, 200× magnification; (d) immunohistochemical staining with S100P, 200× magnification; (e) immunohistochemical staining with γH2AX, 400× magnification; (f) biliary intraepithelial neoplasia, H&E staining, 400× magnification; (g) intraductal papillary neoplasm of the bile duct, 2 mm in diameter, H&E staining, 40× magnification; (h) sclerotic bile duct, H&E staining, 200× magnification, H&E staining; (i) focal bile duct loss (arrow), H&E staining, 100× magnification.

The precancerous/early cancerous lesions could not be pathologically examined in patient 1 because nearly all the biliary epithelium was removed in the operative specimens. IPNB lesions without invasion was detected in three patients (patient 3 had two IPNB lesions, and patients 2 and 5 had one lesion each). Of the four IPNB lesions, two (patients 3 and 5) were classified as micro IPNB (2 mm and 3 mm in diameter, respectively)¹⁶⁾. BilIN lesions were detected in various sites of the large bile ducts²¹⁾ and peribiliary glands in patients 2–5. In all five patients, chronic bile duct injury was observed in various sites of the bile ducts in the noncancerous hepatic tissues. Although fatty liver was detected in patient 4, all five patients did not have cirrhotic changes and other hepatobiliary diseases that are known risk factors for cholangiocarcinoma such as primary sclerosing cholangitis, hepatolithiasis, pancreaticobiliary maljunction, or liver fluke infection.

The results of immunohistochemical studies are shown in Fig. 2 and Table 4. The expression of γ-H2AX was detected in the cholangiocarcinoma tissues (5 of 5 patients with cholangiocarcinoma), BilIN (4 of 4 patients with BilIN), IPNB (3 of 3 patients with IPNB), and non-neoplastic bile duct (in all patients), indicating DNA injury in non-neoplastic bile duct as well as precancerous/early cancerous lesions (BilIN and IPNB) and cancerous tissue. The expression of S100P was detected in cholangiocarcinoma tissues (5 of 5 patients with cholangiocarcinoma), BilIN (4 of 4 patients with BilIN), and IPNB (3 of 3 patients with IPNB), whereas the expression was not detected in non-neoplastic bile duct in all patient, indicating that the malignant transformation has already developed in precancerous/early cancerous lesions as well as cancerous tissue.

Table 4. Expression of γ-H2AX and S100P in the immunohistochemical analysis.

Patient no.	Cholangiocarcinoma	BilIN	IPNB	Non-neoplastic bile duct
				γ-H2AX	S100P
1	Positive	NE	NE	Positive	Negative
2	Positive	Positive	Positive	Positive	Negative
3	Positive	Positive	Positivea	Positive	Negative
4	Positive	Positive	NE	Positive	Negative
5	Positive	Positive	Positive	Positive	Negative

BilIN: biliary intraepithelial neoplasia; IPNB: itraductal papillary neoplasm of the bile duct; NE: BilIN and IPNB could not be evaluated because of the absence of the legions or nearly all the biliary epithelium.

^aPatient 3 had 2 IPNB lesions positive for γ-H2AX and S100P.

Treatment and prognosis

Details of the treatments and prognosis of the five patients are shown in Table 5. Pancreatoduodenectomy with regional lymph node dissection was performed in one patient (patient 1). Liver resection with or without lymph node dissection was performed in patients 3–5. Four patients (patients 1, 3, 4, and 5) were administered adjuvant chemotherapy with S-1 (tegafur/gimeracil/oteracil potassium). The recurrent tumors (liver metastasis in patient 1 and intrabdominal lymph node metastasis in patient 3) were treated successfully by nivolumab therapy²²⁾, according to the supportive evidence^{23, 24)}. In patient 2, chemotherapy with gemcitabine and cisplatin, stenting for the stenosis of the common bile duct, and gastrojejunostomy for duodenal obstruction were performed because the cholangiocarcinoma was detected at the advanced stage (stage IV). The patient died 5 months after the first treatment.

Table 5. Treatments and prognosis of patients with occupational cholangiocarcinoma.

Patient no.	Treatments	Prognosis after the fisrt treatment
1	Pancreaticoduodenectomy with lymph node dissection, adjuvant chemotherapy with S-1, and novolmab for liver metastasis	9 yr, 3 months, alive
2	Chemotherapy with gemcitabine and cisplatin, stenting for the stenosis of the common bile duct, and gastrojejunostomy for duodenal obstruction	5 months, dead
3	Extended left hepatectomy with lymph node dissection, adjuvant chemotherapy with S-1, and novolmab for intraabdominal lymph node metastasis	6 yr, 10 months, alive
4	Anterior segmentectomy of the liver with lymph node dissection and adjuvant chemotherapy with S-1	2 yr, 1 month, alive
5	Extended posterior segmentectomy of the liver with lymph node dissection and adjuvant chemotherapy with S-1	6 months, alive

S-1: tegafur/gimeracil/oteracil potassium.

Exposure to chlorinated organic solvents in the 22 patients

Among 95 workers exposed to DCP at the printing company, cholangiocarcinoma developed in 22 workers (21 men and one woman). Exposure conditions to DCP and DCM are shown in Supplementary Table 1. The age at cholangiocarcinoma diagnosis ranged from 25 to 52 (mean, 39) yr. The duration of exposure to DCP was 6 to 15 (mean, 10) yr and the cumulative exposure was 646 to 3,433 (mean, 2,119) ppm-yr. The period between the first exposure to DCP to the detection of cholangiocarcinoma was 7 to 31 (mean, 17) yr, and the longest period between exposure cessation to the detection of cholangiocarcinoma was 18 yr. Of the 22 patients, 15 were also exposed to DCM. The exposure duration to DCM was 1 to 8 (mean, 5) yr, and the cumulative exposure was 315 to 1,282 (mean, 847) ppm-yr.

The number of patients with occupational cholangiocarcinoma at the end of 2012 were 4 (4 men, 0 women) in 76 workers (60 men, 16 women) for cumulative exposure to DCP of ≤1,499 ppm-yr and 13 (13 men, 0 woman) among 19 workers (18 men, 1 woman) with cumulative exposure to DCP of ≥1,500 ppm-yr. All new five patients with cholangiocarcinoma (4 men, 1 woman) belonged to high exposure group (≥1,500 ppm-yr). Consequently, occupational cholangiocarcinoma developed in 18 (17 men, 1 woman) of the 19 workers (18 men, 1 woman) in the high exposure group at the end of 2023.

Discussion

In this report, we described the subsequent development of occupational cholangiocarcinoma in five patients after the previous report of the 17 patients²⁾ and the characteristics of these five patients with occupational cholangiocarcinoma.

In December 2013, a regular health examination (twice a year), including laboratory test and diagnostic imaging including ultrasonography or CT, for workers exposed to 1,2-dichloropropane for at least 2 yr by the Ministry of Health, Labour and Welfare (health handbook) started. In four of the five patients, cholangiocarcinoma was detected during the regular health examination by the ministry or the company or during the follow-up and treatments for liver dysfunction (high serum γ-GTP activity), and they underwent surgery. Patient 2 visited the hospital because of jaundice and appetite loss; however, curative resection could not be performed because of the advanced stage (stage IV) of the cholangiocarcinoma. The serum γ-GTP activity was elevated in all patients. In patients 3–5 with available laboratory test results from several years before cholangiocarcinoma diagnosis, the serum γ-GTP activities were consistently high or gradually increased, followed by an increase in AST and/or ALT activities. These findings were observed in our previous studies^{2, 8, 9)}. Such an increase in γ-GTP activities must be related to the chronic bile duct injury by DCP and/or DCM. At the time of diagnosis of cholangiocarcinoma, the concentration of CA19-9 was high in three of the five patients. In addition, diagnostic imaging helped in the screening of the other hepatobiliary diseases. Therefore, the combination with laboratory tests and diagnostic imaging was useful for the diagnosis of the cholangiocarcinoma. The longest period between exposure cessation to cholangiocarcinoma detection was 18 yr. These findings indicate the importance of regular health examination comprising the laboratory tests, including AST, ALT, γ-GTP and CA19-9, and diagnostic imaging for the workers exposed to DCP and/or DCM for a long term.

On diagnostic imaging, the main tumor was detected on CT and/or MRI in the five patients. Lymph node metastasis was detected in three patients on diagnostic imaging. FDG-PET showed the main tumor in 3 of the 4 patients who underwent FDG-PET, whereas precancerous/early cancerous lesions were not detected, as previously reported²⁵⁾. On diagnostic imaging, regional dilatation of the intrahepatic bile ducts without tumor-induced obstruction, which was characteristic of occupational cholangiocarcinoma, was observed in 5 of the 17 patients in the previous study and in 3 of the 5 patients (patients 3–5).

Of the 17 patients in the previous report²⁾, 10 had intrahepatic cholangiocarcinoma, 5 had extrahepatic bile duct, and 2 had both intra- and extrahepatic cholangiocarcinoma. Intrahepatic cholangiocarcinoma was classified into the mass-forming type and invasive IPNB (intraductal growth type), and the extrahepatic cholangiocarcinoma was mainly classified as papillary-type. Of the five patients in this report, patients 3–5 had intrahepatic cholangiocarcinoma, one (patient 1) had extrahepatic cholangiocarcinoma, and patient 2 had both intra- and extrahepatic cholangiocarcinoma. Among the four patients with intrahepatic cholangiocarcinoma, the tumors were classified as mass-forming type in patients 3–5 and periductal infiltration type in patient 2. Of the two patients with extrahepatic cholangiocarcinoma, the tumor in patient 1 was classified as flat-infiltration type, whereas the tumor in patient 2 could not be classified because the presence of a bulky mass. The main and most invasive cholangiocarcinoma lesions were located in the common bile duct, hepatic duct, or the first to third branches of the intrahepatic bile duct (large bile duct²⁵⁾), and the intrahepatic cholangiocarcinoma was classified into large duct type according to the WHO classification⁵⁾, similar to the previous studies^{2, 9)}.

In the previous report²⁾, chronic bile duct injury and BilIN were detected in all eight patients whose operative specimens were available for the pathological examination, and IPNB were detected in seven of the eight patients. In this report, all patients had chronic bile duct injury. BilIN was detected in four patients, and IPNB was detected in three of the four patients whose biliary epithelium was available. In addition, the expression of γ-H2AX was detected in the cholangiocarcinoma tissues, BilIN, IPNB, and non-neoplastic bile duct in all available patients, indicating that DNA injury has occurred in non-neoplastic bile duct as well as precancerous/early cancerous lesions (BilIN and IPNB) and cancerous tissue. The expression of S100P expression was detected in cholangiocarcinoma tissues, BilIN, and IPNB in all available patients, indicating that the malignant transformation has occurred in BilIN and IPNB as well as cholangiocarcinoma tissues. Fundamentally, the immunohistochemical results of γ-H2AX and S100P in the five new patients were similar to those in the previous patients. However, the positive ratio of γ-H2AX and S100P appeared to be higher in the present study than that in the previous study. We could investigate the whole operative specimens or autopsy specimens in the five new patients on the basis of the recognition of occupational cholangiocarcinoma (the five patients were treated after 2015), whereas a few specimens for only the diagnosis of cholangiocarcinoma were available in most patients in the previous study. As the results. We could investigate many preinvasive/early cancerous lesions in the present study, resulting in higher positive ratios. High expression of γ-H2AX and S100P indicate that DCP, DCM, or their products initiate a multistep carcinogenesis process by causing DNA damage to the biliary epithelium. This induces bile duct injury and precancerous/preinvasive lesions at various sites of the bile duct, particularly large bile ducts, and invasive cholangiocarcinoma. Chronic bile duct injury was detected at various sites of the bile duct, even at those apart from the main tumor, indicating that the bile duct damage occurred at the entire whole liver. The wide distribution of the bile duct injury with DNA damage and precancerous/preinvasive lesions indicates the highly malignant potential at the various sites of the bile duct, which can lead to multifocal (multicentric) carcinogenesis.

In patient 2, curative treatment was impossible because the cholangiocarcinoma was detected at the advanced stage (stage IV). In the other four patients in whom cholangiocarcinoma was detected during the regular health examination for workers exposed to DCP or the follow-up and treatments for liver dysfunction, pancreatoduodenectomy or liver resection with or without lymph node dissection and adjuvant chemotherapy could be performed. Although cholangiocarcinoma recurred in patients 1 and 3, nivolumab therapy was effective. The four patients are now alive. Performing curative treatment for occupational cholangiocarcinoma appears difficult because cholangiocarcinoma and precancerous/early cancerous lesions (BilIN and IPNB) were often present at various sites of the large bile duct. However, the detection of cholangiocarcinoma at the early stage and the combination of curative resection with adjuvant therapy and alternative treatment must be effective for occupational cholangiocarcinoma.

Epidemiologically, extending the observation period of the cohort study¹⁸⁾ to 2023 would be desirable to show the standardized incidence ratio. In this study, we preliminarily showed the prevalence of cholangiocarcinoma in the high-exposure group (≥1,500 ppm-yr), because we are unable to follow-up everyone at this time. By the end of 2023, cholangiocarcinoma had developed in 18 (17 men, 1 woman) of 19 (95%) workers (18 men, 1 woman) in the high-exposure group, further strengthening the theory that DCP causes occupational cholangiocarcinoma and suggesting the lack of sex difference in the development of occupational cholangiocarcinoma. Therefore, a close follow-up of cholangiocarcinoma is necessary for high-exposure workers.

Of the 22 patients, 15 were also exposed to DCM, so it is possible that DCM may play a certain role in the development of occupational cholangiocarcinoma jointly with DCP. Indeed, DCM is a possible risk factor for biliary tract cancer based on findings from a past epidemiological study²⁶⁾. However, the contribution of DCM could not be determined because exposure levels to DCM correlated with those to DCP in the workers exposed to both chemicals in the printing company. Further epidemiological studies on the role of DCM for cholangiocarcinoma development is necessary.

Other possible factors of the development of occupational include co-exposure to other solvents, underlying risk factors for cholangiocarcinoma, and genetic polymorphisms in the metabolism of DCP and DCM¹⁹⁾. However, this study showed extremely high prevalence of cholangiocarcinoma in the high-exposure group. Because it is unlikely that workers with genes that contribute to the development of cholangiocarcinoma concentrated in the high-exposure group by chance, the role in the genetic polymorphism for cholangiocarcinoma development can be denied. Furthermore, not all the patients were exposed to other chemicals or had known underlying risk factors. Therefore, the results of the previous^{1, 2)} and the present studies indicate the other factors are absent or present in negligible amount compared to DCP.

This study has several limitations. Although the clinicopathological and epidemiological findings of the patients with the occupational cholangiocarcinoma and genetic findings of the cholangiocarcinoma have been investigated, detailed mechanism of the development of the occupational cholangiocarcinoma remain unclear because we could only analyze limited information from the small number of patients with occupational cholangiocarcinoma. We could not evaluate of the laboratory test results in most workers (without cholangiocarcinoma) exposed to DCP and/or DCM at the company because they were already retired from the printing company and could not be contacted. A further study exploring changes in available laboratory test results are necessary.

Conclusions

After the previous report of 17 patients with occupational cholangiocarcinoma, cholangiocarcinoma was newly detected in 5 workers who were exposed to DCP and/or DCM in the same printing company by the end of 2023. The clinicopathological findings in the new 5 patients were similar to those in the previous 17 patients. The prevalence of cholangiocarcinoma in the high-exposure group was extremely high. These findings strengthen further the theory that DCP causes occupational cholangiocarcinoma. Regular health examination or follow up for liver dysfunction of workers exposed to DCP and/or DCM is useful to detect the cholangiocarcinoma at the early stage, and the long-term examination is necessary because the risk for cholangiocarcinoma persists over the long term.

Author Contributions

Study design: S. Kubo, MK, and S. Kumagai. Acquisition of data: S. Kubo, HS, TI, S. Tanaka, S. Maeda, AM, and S. Kumagai. Pathological aspect of the study: MK and YS. Data analysis: S. Kubo, MK, and S. Kumagai. Manuscript draft: S. Kubo, S. Kumagai and GE. All authors reviewed the manuscript.

Funding

This work was supported in part by the Japan Society for the Promotion of Science KAKENHI Grant Number 22K16495 (Novel biomarkers for establishment of personalized therapy based on the tumor microenvironment of intrahepatic cholangiocarcinoma).

Conflict of Interest

The authors declare no conflict of interest for this article.