Article Text
PDF
Abstract
Vanishing bile duct syndrome is an uncommon condition characterised by the progressive loss and disappearance of bile ducts. It is an acquired form of cholestatic liver disease presenting with hepatic ductopenia (loss of >50% bile ducts in the portal areas). We present a case of vanishing bile duct syndrome as a presentation of Hodgkin’s lymphoma who was treated with standard-of-care chemotherapy—doxorubicin, bleomycin, vinblastine and dacarbazine (along with brief administration of rituximab), which led to complete response and normalisation of liver function.
- Oncology
- Medical management
- Drugs and medicines
Statistics from Altmetric.com
Background
Vanishing bile duct syndrome (VBDS) is an uncommon and acquired form of cholestatic liver disease characterised by progressive loss and disappearance of bile ducts resulting in hepatic ductopenia and cholestasis. VBDS may be associated with malignancies like Hodgkin’s lymphoma (HL) and other non-malignancy-related causes including infections like cytomegalovirus (CMV), macrophage activation syndrome (MAS), haemophagocytic lymphohistiocytic syndrome (HLH), autoimmune diseases and drug-related or even idiopathic. In patients with HL, liver infiltration by lymphoma is seen in up to 50% of the patients but progression to VBDS is rare.1–3 Few case reports have been described in this presentation. VBDS is thought to be a paraneoplastic phenomenon and management usually includes the treatment of HL. The standard treatment of HL with doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD) chemotherapy is challenging in cases with liver injury as the chemotherapy regimen itself can be hepatotoxic. The benefit of treating HL associated with VBDS with full-dose ABVD needs to be carefully evaluated against the potential toxicity. We delineate a case of HL with VBDS-induced liver dysfunction who was treated with a full-dose ABVD regimen.
Case presentation
We report a case of a female in her 20s with a medical history of scoliosis who initially presented with the symptoms of nausea, weight loss, right-sided neck mass, pruritus and yellowing of eyes and skin. She was not on any medications except celecoxib for the past 15 days and prednisone 40 mg for the last 5 days due to back pain associated with scoliosis.
On physical exam, she was noted to have jaundice and cervical lymphadenopathy. Laboratory values were significant for hyperbilirubinaemia (total bilirubin 22 mg/dL (normal 0.3–1.2 g/dL)) and elevated liver enzymes (aspartate aminotransferase (AST) 91 U/L (normal 15–41 U/L); alanine aminotransferase (ALT) 108 U/L (normal 7–35 U/L); alkaline phosphatase (ALP) 662 U/L (normal 40–129 U/L)). Imaging with ultrasound of the abdomen showed mild hepatosplenomegaly but no bile duct dilatation. Hepatitis viral panel, Epstein-Barr virus (EBV) and HIV were negative. CA 19-9 and alpha-fetoprotein were within normal limits. Prothrombin time - INR (International Normalized Ratio) was elevated at 1.3. Magnetic resonance cholangiopancreatography (MRCP) was done which did not show any obstructive process/mass in the hepatobiliary region.
Investigations
A liver biopsy was done for her elevated liver enzymes with no other obvious cause. It showed prominent cholestasis and Kupffer cell hyperplasia with focal perivenular hepatocyte dropout and loss of bile ducts, >50%. This was consistent with idiopathic cholestasis with ductopenia, also known as VBDS (figure 1A–D).
(A) Lymph node biopsy, H&E stain, 200×. (B) Lymph node biopsy, CD30 immunohistochemical (IHC) stain, 100× (highlights CD30+ Reed-Sternberg and Hodgkin cells). (C) Liver biopsy, H&E stain, 100× (shows bile duct loss). (D) Liver biopsy, CK7 IHC, 200× (shows vanishing bile ducts, only isolated cells CK7+ with no intact bile ducts).
Initial positron emission tomography (PET)/CT of the neck showed an extensive hypermetabolic supradiaphragmatic lymphadenopathy, including a large conglomerate nodal mass within the right neck and superior mediastinum with associated mass-effect and infiltration of the brachial plexus. Furthermore, it showed left internal mammary disease which extended through the chest wall and involved the sternum. This finding was discussed at the tumour board of our institution, and it was concluded that the involvement of the sternum was equivocal with the lymph node abutting the sternum rather than actual bone involvement.
The cervical lymph node core biopsy was consistent with classical HL, nodular sclerosis type and the stage was decided to be stage II unfavourable. Immunohistochemical staining performed on block A1 showed that the tumour cells were positive for CD30 (Fig 1B) and CD15 and negative for CD3, CD5, Pan-Mel (HMB-45, MART-1), pan-CK, CD10, ALK and CD45 supporting the diagnosis. Bone marrow biopsy was performed which was negative for any involvement by HL.
Differential diagnosis
Our patient presented with signs and symptoms of liver dysfunction and lymphadenopathy. Liver enzymes showed a cholestatic pattern. The differential diagnosis included infectious causes like viral hepatitis, autoimmune diseases including primary biliary cholangitis (PBC), especially given female gender, primary sclerosing cholangitis (PSC) or malignancy including primary liver cancer or more likely lymphoma given the patient’s relatively young age. We obtained hepatitis viral panel, EBV and HIV which were negative. The patient was not immunocompromised and so further viral testing (including CMV) was not performed. Autoimmune serological workup was unremarkable. This included mitochondrial antibody (for PBC), IgG subclasses (for PSC), antinuclear antibody (ANA for lupus) and smooth muscle antibody (autoimmune hepatitis). MRCP did not show any obstructive lesion as well as suspicious findings for PBC or PSC. A liver biopsy was then pursued which revealed ductopenia consistent with VBDS. This could be seen with infections, autoimmune disease, MAS, HLH, graft versus host disease (GVHD) or malignancies like lymphoma. MAS or HLH usually have more severe clinical presentations. Furthermore, these conditions are often associated with extremely high ferritin levels which was not the case in our patient. Interestingly, celecoxib has been shown to be associated with VBDS and our patient has been taking it.4 The patient had no history of transplant and so GVHD was not considered. However, the patient had another significant finding of cervical lymphadenopathy. Cervical lymph node biopsy was pursued simultaneously, and it clinched the diagnosis of HL. PET/CT scan showed extensive supradiaphragmatic lymphadenopathy. HL has been known to cause VBDS and thus in our patient VBDS was attributed to HL. If no cause is discovered after exhaustive workup, idiopathic ductopenia is considered.1 2
Treatment
The primary goal of treatment for her hepatic injury was treatment of the HL with close monitoring of her liver dysfunction. She was initially started on gemcitabine, cyclophosphamide and dexamethasone as liver-sparing agents. Her liver enzymes (AST, ALT) improved but total bilirubin continued to rise. After the literature review, a decision was made to start chemotherapy with a full dose of ABVD. Rituximab was added as there have been some case reports describing activity against autoimmune bile duct injury.1 2 5 6 She was also started on ursodeoxycholic acid (UDCA) 250 mg daily as it is a typical pharmacologic agent used for severe hepatic ductopenia. UDCA has been shown to modulate hepatobiliary secretion by transcriptional and post-transcriptional mechanisms as well as block apoptosis in hepatocytes by interrupting the classic apoptotic pathways. Furthermore, UDCA inhibits hydrophobic bile acid-induced membrane disruption mainly by alteration of the structure and composition of simple and mixed micelles. Thus, UDCA reduces periportal inflammation, enhances ductular proliferation and delays the progression of fibrosis.7
1 month later, she received a second cycle of ABVD-R (Doxorubicin, Bleomycin, Dose-adjusted Vinblastine, Dacarbazine and Rituximab). PET/CT imaging after the second cycle showed a significant decrease in the overall size of the supraclavicular mass measuring approximately 4.5×2.2 cm, a significant interval decrease in the metabolic activity of this lesion and a complete interval resolution of the mediastinal lymphadenopathy. The Deauville score was 2. Further cycles of chemotherapy included cycles 3–6 of ABVD each, 1 month apart. After a total of five cycles, PET/CT findings were consistent with a complete metabolic response, and the Deauville score was 1. With these imaging findings, it was decided to proceed with two additional cycles of ABVD to complete a total of six cycles to avoid radiation to the chest given her young age. She received a total of seven chemotherapy cycles (one cycle of liver-sparing chemotherapy and six cycles of ABVD; the first two cycles of ABVD had R with adjusted vinblastine). After seven cycles, the appearance of the soft tissue infiltrative mass-like process in the right supra and subclavicular regions was unchanged, and Standardized Uptake Value (SUV) was 1.8 with no significant change. The appearance of radiotracer activity in the bilateral submandibular glands was also similar. Thus, the patient had a complete response (CR) with no evidence of disease based on PET-CT scan. Furthermore, her liver dysfunction resolved progressively with chemotherapy (figures 2 and 3).
Pattern of liver enzymes. ABVD-R, doxorubicin, bleomycin, vinblastine, dacarbazine and rituximab; ABVD, doxorubicin, bleomycin, vinblastine and dacarbazine; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase.
Pattern of total bilirubin. ABVD-R, doxorubicin, bleomycin, vinblastine, dacarbazine and rituximab; ABVD, doxorubicin, bleomycin, vinblastine and dacarbazine.
Outcome and follow-up
The patient continues to follow-up with haematology and oncology clinic every 6 months and is doing well. She did have a CT scan at a 2-year follow-up, and imaging was consistent with unchanged ill-defined soft tissue thickening/induration centred in the right supraclavicular soft tissues, without any evidence of disease progression in the chest or abdomen/pelvis. Liver enzymes normalised to AST 36 U/L, ALT 56 U/L and ALP 110 U/L and total bilirubin 0.8 mg/dL. A repeat liver biopsy to look for the resolution of ductopenia was not pursued given the risks of a biopsy and the patient being in complete radiological and clinical remission.
Discussion
VBDS is an uncommon and acquired form of cholestatic liver disease characterised by hepatic ductopenia and cholestatic pattern of liver injury. This cholestatic liver disease may also progress to biliary cirrhosis. Ductopenia is defined as loss of the intralobular bile ducts in >50% of small portal tracts. A minimum of 10 portal tracts have to be visualised in the histological specimen.2
Many causes of VBDS have been identified.8 It can be a consequence of drug-induced liver injury, commonly resolving after the offending agent is discontinued.9 Some other causes include immune-mediated disorders like PBC, PSC, chronic GVHD and sarcoidosis. It is also caused by viral infections like hepatitis B, hepatitis C, EBV and CMV (in immunocompromised patients). Other conditions associated with VBDS include MAS, HLH and malignancies like HL.10 The normalisation of liver function with the treatment of HL confirmed the association of VBDS with HL in our patient.
In patients presenting with direct hyperbilirubinaemia and cholestatic pattern of liver enzymes, the diagnosis of VBDS related to other causes requires a high index of suspicion. A liver biopsy is pursued after ruling out the causes of extrahepatic cholestasis usually by Endoscopic retrograde cholangiopancreaticography (ERCP) or MRCP, and intrahepatic cholestasis caused by autoimmune causes, PBC and PSC. General measures for the treatment of VBDS include avoiding drugs associated with liver injury, abstinence from alcohol, lifestyle modifications to maintain ideal body weight (including dietary therapy and exercise) and vaccinations against hepatitis A and B.
In HL, the common causes of abnormal liver function are extrahepatic bile duct obstruction, direct hepatic involvement by lymphoma cells, viral hepatitis or complications of chemotherapy. Direct liver infiltration by lymphoma is seen in up to 50% of the patients but progression to VBDS is rare.3 11 Cholestasis with bile duct loss is a well-known but rare presentation of HL.9
The proposed mechanisms causing ductopenia include cell-mediated immunological reactions or toxic cytokines derived from lymphoma cells.12 Alternative theories include damage to biliary epithelium due to cross-reactive T-cells recognising tumour-associated antigens expressed on biliary cells causing apoptosis.12 13 Moreover, some studies have shown that the cytokines produced by HL induce the expression of major histocompatibility complex and intercellular adhesion molecules.14
Previous case reports have been described with idiopathic cholestasis as a presentation and management has included the treatment of HL. Patients with persistent cholestasis may be evaluated for liver transplantation.15 The standard treatment for HL with the lowest relapse rate consists of ABVD with the number of cycles determined by the stage of HL. For stage II unfavourable HL, we typically do three to four cycles of ABVD. This is followed by involved-site radiation therapy (ISRT) in select circumstances. The intensity of dosing directly affects the outcome of HL.16 17 However, young women are at a high risk of developing complications related to radiation therapy, including breast cancer after chest irradiation. This was taken into consideration in our patient as well. Treatment for VBDS in the literature has been variable, although the focus usually lies on treating the underlying disease. In our patient, PET showed Deauville 2 after two cycles and CR after four cycles. The treatment choices were discussed—ISRT versus completion of additional chemotherapy. Our initial plan was treatment with four cycles of ABVD followed by ISRT (given stage II unfavourable disease, >3 sites and bulky disease), however, given young age there was a consensus to avoid ISRT. Bleomycin was continued and not de-escalated as per RATHL (response-adapted therapy for advanced Hodgkin lymphoma) protocol to avoid ISRT in a young female and also based upon initial scans there was a concern for more advanced disease.18 There were also significant insurance and financial barriers to escalating treatment to autologous hematopoetic stem cell transplant if required.
Generally, many chemotherapy drugs cause hepatocellular injury, inflammation and/or cholestasis, while some other drugs cause endothelial damage and thrombosis leading to vascular complications such as hepatic veno-occlusive disease (hepatic sinusoidal obstruction syndrome).19 Cases of fatal outcomes like necrosis of the liver and acute hepatic vein thrombosis with ABVD have been reported.20 As per previously published cases of HL-related VBDS, the 1-year overall and non-liver failure survival is 43% and 41%, respectively.21
If liver dysfunction in HL is determined primarily to be caused by VBDS, the implementation of liver sparing cycle of chemotherapy may not be necessary, and its significance is unknown. As evidenced in our case, the treatment of VBDS with a standard ABVD regimen, along with the treatment with UDCA and rituximab, led to disease remission as well as hepatic recovery.
The role of rituximab, a chimeric monoclonal antibody against CD20 (which is commonly negative in classic HL), has been explored. Rituximab may have a therapeutic role even in CD20-negative HL patients given the presumed immunological reaction causing VBDS. Apart from chemotherapy and radiotherapy, apheresis treatment has also been reported to be effective in achieving a cure.22
More recently, the role of brentuximab vedotin (BV) has been explored in cases of liver dysfunction in HL. Of the multiple cases described in the literature, complete remission was obtained in two cases. Various dose adjustments have been described, but it might be concluded that a reduced dose of BV at 1.2 mg/kg could be an effective and relatively safe treatment option for patients with HL and VBDS.20 21 23 Brentuximab has replaced bleomycin in the ABVD regimen now. A+AVD (Brentuximab vedotin, doxorubicin, vinblastine and dacarbazine) is also a preferred first-line treatment option for advanced HL.24–26
As A+AVD therapy is approved only for stage III and IV HL, it was not used in our patients. The ECHELON-1 trial only included patients with total bilirubin level <1.5 times the upper limit of normal (except for patients with Gilbert’s syndrome) and ALT or AST level <3 times the upper limit of normal (except for patients with involvement of the liver). Additionally, the frequency of grade 3 adverse reactions (peripheral neuropathy, infectious colitis, paralytic ileus, unspecified shock and CMV infection), and deaths was greater in patients with moderate and severe hepatic impairment compared with patients with normal hepatic function.3 26 27
When the pharmacokinetics of BV and monomethyl auristatin E (MMAE) were evaluated, it was found that the AUC (area under the curve) of MMAE was almost 2–3 times higher in patients with hepatic impairment when compared with the patients with normal hepatic function. It is recommended to use a reduced dosage of BV in mild hepatic impairment (Child-Pugh A) and avoid use in moderate (Child-Pugh B) and severe hepatic impairment (Child-Pugh C).28
ABVD therapy has been approved for many decades with well-known pharmacokinetics, and there have been multiple trials for early-stage disease regardless of favourable or unfavourable outcomes. Given the extensive historical experience with ABVD-R and its well-established pharmacokinetic profile, it emerges as a more favourable therapeutic option for our patient. There is also no dosage adjustment required for rituximab for hepatic impairment, and its pharmacokinetics are well-established. Some caveats of this therapy include the known potential hepatotoxicity of the ABVD regimen as well as the hepatic metabolism required to convert these drugs to their active metabolites. The exclusion of liver disease from these trials as a factor further complicates the ability to anticipate the advantages of A+AVD over ABVD within the context of our patient’s treatment.29–31
In conclusion, in idiopathic cholestasis, a high clinical suspicion of the underlying malignancy must be present. VBDS is a rare cause of abnormal liver function in HL. Early and complete treatment of HL with full-dose ABVD therapy has previously been shown to achieve remission as well as resolution of liver dysfunction. Our case also shows similar complete remission in VBDS associated with underlying HL with treatment by full-dose ABVD.
Patient’s perspective
I was diagnosed with Lymphoma Hodgkin’s several years ago. My symptoms started about a year prior to my diagnosis. I was having recurring pain in the right side of the neck and a lump on it. In the beginning, I thought I had pulled a muscle in the gym. I went to different doctors, and they prescribed me steroids since that was the only symptom I had, and the pain felt like it was muscular. After taking steroids and acetaminophen, the pain would go away but come back, and this was a cycle. It started getting swollen more every day and I started having other symptoms such as vomiting and weakness, additionally, my skin and eyes started turning yellow. I decided to visit a more specialised pain management doctor since there was obviously something going on and he ordered a CT scan of my neck, that is when they realised the illness. It was tough at the beginning because everything changed from 1 day to another, and the doctors were concerned about my liver since it was being impacted. I was told that I needed to start chemotherapy as soon as possible, however, the liver function was a challenge. For this reason, they decided to start with a low dose of treatment to see what the reaction was, fortunately, my liver started responding and my enzyme levels as well as bilirubin started slowly going down.
The whole process was frightening at the beginning, but I am blessed to have had the support of my family and friends, and of course, of my doctors that decided on the right treatment for me and helped me get through it. I had a positive and good attitude during my treatment, I used to meditate a lot and exercise as much as I could to keep myself active. After 2 months of the treatment, my skin colour was almost back to normal, and my liver enzymes are still going back to normal after almost 3 years. After my treatment was completed, I have had emotional ups and downs, but I have been doing very well overall after 3 years.
Learning points
Vanishing bile duct syndrome is a rare cause of abnormal liver function in Hodgkin’s lymphoma (HL).
In patients with idiopathic cholestasis, a high clinical suspicion for the underlying malignancy must be present, after ruling out other causes.
Early and complete treatment of HL with full-dose doxorubicin, bleomycin, vinblastine and dacarbazine (ABVD) therapy achieves remission as well as resolution of the liver dysfunction.
Consider treatment of HL with full-dose ABVD even with impaired liver function.
Ethics statements
Patient consent for publication
Acknowledgments
Valerica Mateescu, MD was responsible for sourcing pathology images.
References
Footnotes
X @ifrahfatima
Contributors The following authors were responsible for drafting of the text, sourcing and editing of clinical images, investigation results, drawing original diagrams and algorithms, and critical revision for important intellectual content: IF, HM, SM and AS. The following authors gave final approval of the manuscript: IF, HM, SM and AS.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.