ORIGINAL RESEARCH

Understanding Patient Experience in Biliary TractCancer: A Qualitative Patient Interview Study

Nikunj Patel . Xandra Lie . Chad Gwaltney . Nana Rokutanda .

Afsaneh Barzi . Davide Melisi . Teresa Macarulla . Makoto Ueno .

Seung Tae Kim . Oren Meyers . Christina Workman . Melinda Bachini .

Gordon Cohen

Received: April 15, 2021 /Accepted: June 10, 2021 / Published online: July 10, 2021� The Author(s) 2021

ABSTRACT

Introduction: Patients living with biliary tractcancer (BTC) experience a decline in health-related quality of life (HRQoL). This studyaimed to obtain a comprehensive

understanding of the patient experience of BTC-related signs/symptoms and the impacts ofthese on daily functioning and HRQoL.Methods: Patients with BTC participated inqualitative semi-structured concept elicitationinterviews. Signs/symptoms and impacts of BTCwere initially explored by targeted literaturesearches and interviews with five clinicians.Patient interviews were transcribed and codedusing qualitative research software. Conceptsaturation was assessed over five interviewwaves. A sign/symptom or impact was defined

Supplementary Information The online versioncontains supplementary material available at https://doi.org/10.1007/s40487-021-00159-z.

N. Patel (&)AstraZeneca, 1 Medimmune Way, Gaithersburg, MD20878, USAe-mail: nikunj.patel@astrazeneca.com

N. Rokutanda � C. Workman � G. CohenAstraZeneca, Gaithersburg, MD, USA

X. LieIQVIA, Amsterdam, The Netherlands

C. GwaltneyGwaltney Consulting, Westerly, RI, USA

A. BarziCity of Hope Comprehensive Cancer Center,Duarte, CA, USA

D. MelisiDigestive Molecular Clinical Oncology Unit,Universita degli Studi di Verona, Verona, Italy

D. MelisiExperimental Cancer Medicine Unit, AziendaOspedaliera Integrata di Verona, Verona, Italy

T. MacarullaVall d’Hebron University Hospital, Barcelona, Spain

T. MacarullaVall d’Hebron Institute of Oncology, Barcelona,Spain

M. UenoDepartment of Gastroenterology, Kanagawa CancerCenter, Yokohama, Japan

S. T. KimSamsung Medical Center, SungkyunkwanUniversity School of Medicine, Seoul, South Korea

O. MeyersIQVIA, New York, NY, USA

M. BachiniCholangiocarcinoma Foundation, Riverton, UT,USA

Oncol Ther (2021) 9:557–573

https://doi.org/10.1007/s40487-021-00159-z

as ‘‘salient’’ if mentioned by C 50% of patients,with a mean disturbance rating of C 5 (0–10scale). A conceptual model of the patientexperience of BTC-related signs/symptoms andimpacts was produced.Results: Twenty-three patients from the USA(78% women; median age: 54 years), diagnosedas having early (n = 3), locally advanced(n = 11) or metastatic (n = 9) disease, wereinterviewed. Sixty-six signs/symptoms and 12impacts were identified. Of these, 46 signs/symptoms and 8 impacts were not identifiedfrom the targeted literature or clinician inter-views. Concept saturation was reached by thefourth of five interview waves. Fourteen disease-related signs/symptoms (including fatigue/lackof energy, abdominal pain, lack of appetite,insomnia and diarrhoea) and three impacts(physical, emotional and cognitive impacts)were deemed ‘‘salient’’. The conceptual modelincluded 50 signs/symptoms and 12 impacts.Conclusion: Patients with BTC reported a rangeof signs/symptoms and impacts that negativelyaffect daily functioning and HRQoL.

Keywords: Biliary tract cancer; Interview study;Qualitative research

Key Summary Points

Why carry out this study?

Patients living with biliary tract cancer(BTC) experience a decline in health-related quality of life (HRQoL).

This study aimed to obtain acomprehensive understanding of thepatient experience of BTC-related signs/symptoms and the impacts of these ondaily functioning and HRQoL.

What was learned from the study?

Qualitative interviews in patientsdiagnosed with BTC identified signs/symptoms and impacts associated withthe disease experience, including somethat may not have been previouslyidentified by the literature and cliniciansexperienced in treating BTC.

Understanding patient experience caninform the selection of patient-reportedoutcomes in clinical trials of BTCtreatments.

DIGITAL FEATURES

This article is published with digital features,including a summary slide, to facilitate under-standing of the article. To view digital featuresfor this article go to https://doi.org/10.6084/m9.figshare.14762973.

INTRODUCTION

Biliary tract cancer (BTC; also known as bileduct cancer) includes malignancy of the peri-hilar, distal and intrahepatic bile ducts and gallbladder [1]. While the incidence of BTC is lowin Western countries (0.35–2 cases per 100,000population), studies report that the frequencyof cases has increased progressively over thepast four decades [2, 3]. The incidence of BTC isgenerally higher in Asian and South Americancountries when compared with the rest of theworld [4]. In Northeast Thailand, BTC is themost common type of cancer and, accordingly,the region has the highest prevalence of BTCworldwide ([80 cases per 100,000 population)[5–7]. With the exception of the earlier stages ofgall bladder cancer, which have 5-year survivalrates of between 50% and 80%, the prognosisfor BTC is poor across all stages of disease, with

558 Oncol Ther (2021) 9:557–573

5-year survival rates of 5–30% [4]. One factorthat is likely to contribute to these poor survivalrates is that BTCs are often diagnosed at aninoperable or even at a metastatic stage [8–10].Further, despite potentially curative surgicalremoval for localized disease, relapse rates arehigh [11]. For patients with late-stage disease,palliative treatment may be the only option,usually in the form of chemotherapy.

Patients with BTC have reduced health-re-lated quality of life (HRQoL) due to a combi-nation of tumour- and treatment-related signsor symptoms (signs/symptoms) and the impactof these signs/symptoms on functioning.Tumour-related signs/symptoms vary depend-ing on tumour type, location and stage of thedisease [12]. Signs/symptoms reported in theliterature include jaundice/yellow skin colour,pain/discomfort in stomach area, back pain,other unspecified pain, nausea, fatigue/lack ofenergy, weight loss, sleep problems, lack ofappetite and general unwell feeling [13–18].Impacts mentioned in the literature includeanxiety, inability to do usual activities, depres-sive mood, trouble meeting the needs of thefamily and financial difficulties [13, 16, 18].Patients’ HRQoL tends to decline as the diseasebecomes more advanced. This may, in part, beassociated with patients undergoing moreinvasive surgical procedures, systemicchemotherapeutic treatments and/or palliativetreatments during the later stages of BTC[19, 20].

Patient-reported outcome (PRO) question-naires are increasingly being utilized to assessoutcomes in cancer and are often recommendedby regulatory agencies to evaluate the impact ofthe disease and treatment on patient-reportedsymptoms, functioning and HRQoL [21–23]. Inthe development of a new PRO questionnaire orthe assessment of whether an existing ques-tionnaire is fit for purpose, the concepts thatneed to be covered are identified from thepublished literature, interviews with cliniciansexperienced in treating the target patient pop-ulation and, most importantly, interviews withpatients [24]. Concept elicitation interviewswith patients can capture in-depth qualitativeinformation of the experience of patients livingwith a disease and can be used to assess whether

a PRO questionnaire effectively captures rele-vant concepts of the patient experience, such asdisease- and treatment-related symptoms.

There is a lack of data on the lived experienceof patients with BTC, particularly studies thatconsider the patient experience at differentdisease stages; the limited studies availablefocus on the patient experience of specificoncological treatments [25, 26] or patientinvolvement in care [27]. This lack of data maybe reflective of the relative rarity of BTCs. Theaim of the current study was to obtain a com-prehensive understanding of the patient expe-rience of BTC-related signs/symptoms andimpacts on daily functioning and HRQoL, atdifferent stages of disease and for various BTCsubtypes, to guide patient-centred outcomesmeasurement.

METHODS

This was a qualitative patient interview study,which led to the development of a conceptualmodel of the patient experience of BTC. Aninitial literature search (identifying BTC-relatedsigns/symptoms, impacts and existing PROquestionnaires) and qualitative interviews withclinicians were also performed to guide thepatient interviews. The qualitative interview,research protocol, interview guide and allpatient communication documents werereviewed and approved by the New EnglandInstitutional Review Board (NEIRB) in January2019.

Targeted Literature Search

A preliminary all-stage conceptual model ofdisease-related signs/symptoms and impacts inpatients with BTC was developed from targetedliterature searches in PubMed of publishedarticles from 1 January 2013 to 16 August 2018.This data search was used to inform a clinicaltrial starting 16 April 2019. Search terms inclu-ded BTC disease-related, patient experience andPRO terms (Supplementary Table 1).

Oncol Ther (2021) 9:557–573 559

Qualitative Interviews

Clinician InterviewsSemi-structured interviews were conducted overthe telephone with clinicians experienced intreating BTC in Italy, Japan, Spain, South Koreaand the USA. Interviews lasted approximately75 min and were conducted by three trainedinterviewers (one male and two female researchconsultants, educated to MSc or PhD level,including author Xandra Lie) experienced inqualitative research across a wide range oftherapeutic areas. Interviewers were withoutbias or assumptions regarding this study. Therewas no relationship between interviewers andclinicians prior to study commencement. Theinterviews were conducted using a standardizedinterview guide with open-ended and promptedquestions to explore the symptoms and impactsthat the clinicians observed in patients withBTC. Clinicians reviewed the concepts identi-fied from the review of the literature.

Patient InterviewsTwenty-three patients from the USA wererecruited for the interview study with assistancefrom the patient advocacy group, the Cholan-giocarcinoma Foundation (CCF) (https://cholangiocarcinoma.org/). Participants pro-vided consent online and then answered severalscreening questions to confirm eligibility. Eli-gibility criteria included being C 18 years oldwith a diagnosis of BTC, including cholangio-carcinoma (perihilar, distal or intrahepatic)and/or gall bladder carcinoma, based on histo-logical/radiological confirmation and physiciandiagnosis confirmation forms. Patients withearly (stage I, gall bladder and perihilar BTC, ordistal or intrahepatic BTC), locally advanced(stage II–IVA, gall bladder or perihilar BTC, orstage II–III, distal or intrahepatic BTC) andmetastatic (stage IVB, gall bladder or perihilarBTC, or stage IV, distal or intrahepatic BTC)disease were included in this study. All 23recruited patients participated in this study.

Patient Interview ProcedureA semi-structured interview guide was devel-oped, informed by data gathered from the

qualitative literature review and clinicianinterviews. The telephone interviews were con-ducted by three trained interviewers (one maleand two female research consultants, educatedto MSc or PhD level) with experience in con-ducting individual patient concept elicitationinterviews, and lasted between 75 and 90 mineach; other project team members were occa-sionally present on the calls for training pur-poses (with permission from the patient). Therewas no relationship between interviewers andpatients prior to study commencement. Partic-ipants were made aware that the output of theinterview may contribute to the development ofnew BTC treatments. Participants were asked aset of open-ended questions, and interviewersfollowed up with probing questions as neededto explore the patient experience of conceptsthat patients did not mention spontaneously.Patients were asked about their first experienceof the condition and how their experience mayhave changed over time, and were asked to listand discuss the current signs/symptoms andimpacts of the condition and its treatments. Tofurther explore the underlying causes of symp-toms, during the interview patients were askedwhether they thought symptoms were ‘‘disease-related’’, ‘‘treatment-related’’ or ‘‘disease- andtreatment-related’’, as well as whether a symp-tom was experienced before, during or aftertreatment. No further interviews were con-ducted, and patients did not provide feedbackon the findings. The interview methodologyutilized in this study is in line with recom-mendations provided by the InternationalSociety for Pharmacoeconomics and OutcomesResearch (ISPOR) Good Research Practices TaskForce [24].

Data Analysis

The subtype of participants’ BTC (intrahepatic,perihilar or gall bladder cancer) was recorded toenable sub-group analysis. Prior to descriptivecoding, patient interviews were recorded, tran-scribed and anonymized for thematic analysis.The interview transcripts were then coded,using a codebook developed from the prelimi-nary conceptual model using qualitative

560 Oncol Ther (2021) 9:557–573

research software (Atlas.ti v8, Atlas.ti ScientificSoftware Development GmbH, Germany). Tworesearchers coded the same transcripts inde-pendently, testing for inter-coder agreementafter each transcript and discussing changes tothe codebook and coding rules. After threetranscripts, the coders achieved good inter-coder agreement (predefined as Krippendorff’sC-alpha binary[ 0.7 [28]), and the remainingtranscripts were divided among the coders.During coding, spot checks were conducted toensure the codes made sense based on patientresponse and code language, and groupingswere refined as needed. The frequency of con-cepts was cross-checked with live capture sheetsthat were filled in during the interviews by theinterviewers. Concepts were deemed ‘‘salient’’ ifC 50% of patients mentioned the concept and ithad a disturbance rating of C 5 on a scale of0–10, where 0 is ‘‘not disturbing’’ and 10 is ‘‘verydisturbing’’.

Concept Saturation

The principle of concept saturation was used toassess the adequacy of the sample size [24]. Asample size of 12–25 participants is optimal forconcept elicitation interviews and for reachingsaturation of concept [29–31]. Transcripts weregrouped chronologically into five groups (fourgroups consisting of five interviews and onegroup with three interviews). To evaluate con-cept saturation, the concepts derived from eachgroup were compared with concepts from pre-vious groups to determine whether any newconcepts were present. If new conceptsappeared in the transcripts from the next group,saturation had not yet been achieved. Datasaturation was not discussed with patients.

All-Stage Conceptual Model

Signs/symptoms or impacts related to BTC wereused to construct a model of concepts for thepatient experience of early, locally advancedand metastatic disease stages and BTC subtypes.An initial conceptual model was created fromthe reviews of the literature and was refined andfinalized based on the qualitative clinician and

patient interviews. The final conceptual modelhighlighted the concepts that were consideredsalient. Signs/symptoms or impacts deemed bypatients and clinicians to be exclusively treat-ment-related were not included in the concep-tual model.

RESULTS

Targeted Literature Search

No recent BTC-related qualitative researchstudies were identified from the literaturereview. Six non-qualitative studies were identi-fied (two retrospective cohort studies and fourprospective studies) [14, 32–36], and 19 con-cepts related to signs/symptoms or impacts ofBTC were extracted from these and included inthe preliminary conceptual model. These stud-ies used validated questionnaires to assesspatients’ HRQoL in the physical, mental andsocial domains in the context of BTC[14, 32–36]. The preliminary conceptual modeldeveloped from the literature was used to helpguide the qualitative interviews.

Qualitative Interviews

Clinician InterviewsFive clinicians, specialized in hepato-oncologyand with experience in treating patients withBTC, prioritized four (out of 14) signs/symp-toms of BTC (abdominal pain, lack of appetite,fatigue/lack of energy, pruritus/itching) and two(out of five) disease-related impacts (decreasedphysical functioning and insomnia) identifiedin the literature searches. Based on the clinicianinterviews, several key changes were made tothe preliminary conceptual model. Constipa-tion and diarrhoea were removed because theywere considered treatment-related. Revisedgroupings were made for (1) nausea/queasiness;(2) itching; (3) jaundice, yellow skin/yelloweyes/changes in urine; and (4) fever/chills.Based on the clinician interviews, additionalsymptoms, including vomiting, difficulty eat-ing/feeling of fullness and muscle loss, wereidentified.

Oncol Ther (2021) 9:557–573 561

Patient InterviewsA total of 23 patients from the USA who hadBTC were interviewed (78% women; medianage: 54 years [range 27–80]). Patients werediagnosed as having early (n = 3), locallyadvanced (n = 11) or metastatic (unresectable;n = 9) disease and with BTC subtypes of gallbladder cancer (n = 1), intrahepatic (n = 17) or

perihilar (n = 5). Patient demographics andclinical characteristics are summarized inTable 1. Procedures and pharmaceutical thera-pies received as part of BTC treatment aresummarized in Supplementary Table 2. Patientinterviews identified 78 concepts (66 signs/symptoms and 12 impacts). For some inter-views, insufficient time was available to discussall disturbance ratings. Consequently, thenumber of patients who mentioned a distur-bance rating for a sign/symptom or impact doesnot always equal the number of patients whomentioned the sign/symptom or impact (Figs. 1and 2). Of these, 46 signs/symptoms and eightimpacts had not been identified from the tar-geted literature searches or clinician interviews.

Signs/symptoms not previously identifiedfrom the targeted literature search and clinicianinterviews were related to pain and discomfort(e.g. stomach pain, dull ache, tenderness), gas-trointestinal (GI) signs/symptoms (intestinalgas), strength (muscle tension, weakness),weight changes (weight gain), hair (hair loss,hair change), pain and discomfort in other partsof the body (headaches, head/ear pressurechanges), skin (nail changes, rash), generalunwell feeling (flu-like symptoms, aching allover) and urogenital (dark urine, urinary tractinfection, urgency to urinate). Signs/symptomsrelated to circulatory issues (e.g. sepsis, bloodclot), throat/mouth signs/symptoms (e.g. drymouth, cannot drink fluid), the senses (changein taste, problem with hearing, sensitivity tonoise), the eyes (vision changes, dry eyes) andextremities (hand and foot syndrome, swollenankles/arms/legs, neuropathy) were not identi-fied in the literature search or mentioned by theclinicians. Several signs/symptoms and impactsidentified in the patient interviews were varia-tions of concepts previously identified in theliterature search and clinician interviews. In theliterature search, insomnia was associated onlywith itching; however, patients mostly men-tioned insomnia in association with worries andanxieties, aching and pain.

Across the three disease stages, 14 disease-related signs/symptoms (Table 2) and twotreatment-related signs/symptoms were deemedsalient. Fatigue/lack of energy was reported byall patients, with the highest mean disturbance

Table 1 Demographic and clinical characteristics ofpatients participating in qualitative interviews

Demographic characteristic Patients (N = 23)

Sex, n (%)

Female 18 (78)

Male 5 (22)

Age, years

Mean (SD) 55 (12.8)

Median (range) 54 (27–80)

Tumour status at time of interview, n (%)

Earlya 3 (13)

Locally advancedb 11 (48)

Metastaticc 9 (39)

Ethnicity, n (%)

Caucasian 23 (100)

Disease subtype, n (%)

Perihilar 5 (22)

Intrahepatic 17 (74)

Gall bladder 1 (4)

Education, n (%)

Bachelor’s/graduate degree 14 (61)

Some years of college education 4 (17)

High school 5 (22)

BTC biliary tract cancer, SD standard deviationa Stage I, gall bladder and perihilar BTC, or distal orintrahepatic BTCb Stage II–IVA, gall bladder or perihilar BTC, or stageII–III, distal or intrahepatic BTCc Stage IVB, gall bladder or perihilar BTC, or stage IV,distal or intrahepatic BTC

562 Oncol Ther (2021) 9:557–573

Oncol Ther (2021) 9:557–573 563

rating (7.9, the mean of 21 patients’ ratings) ofall signs/symptoms. Abdominal pain was thesecond most commonly reported symptom(mentioned by 19 patients), with a high meandisturbance rating of 7.6 (the mean of 16patients’ ratings). Other salient disease-relatedsigns/symptoms were lack of appetite, difficultyeating/feeling of fullness, abdominal bloating,diarrhoea, nausea/queasiness, constipation,insomnia, other pain, itchy skin, fever/chills,sensitivity to cold and muscle loss. Patients alsodeemed three impacts to be salient (Table 3);these were physical impacts (difficulty walking)mentioned by 22 patients, with a mean distur-bance rating of 7.5 (the mean of 15 patients’

ratings), emotional impacts (depression) men-tioned by 17 patients, with a mean disturbancerating of 8.5 (the mean of 15 patients’ ratings),and cognitive impacts (memory loss, fuzzybrain) mentioned by 14 patients, with a meandisturbance rating of 8.2 (the mean of 15patients’ ratings).

Signs/Symptoms and Impacts Across DiseaseSubtypesThe following signs/symptoms were mentionedby patients across all disease subtypes of intra-hepatic, perihilar and gall bladder: fatigue/lackof energy, insomnia, dry mouth, cough/throatirritation/voice change, hair loss, abdominalpain, other pain, back pain, sensitivity to cold,change in taste, muscle loss or weight loss. Thesalient sign/symptom of itchy skin was men-tioned by 10 of 17 patients with the intrahep-atic subtype and four of five patients with theperihilar subtype, but not by the patient withthe gall bladder subtype. Emotional and cogni-tive impacts and experiencing difficulties inmeeting the needs of family were impactsmentioned by all patients. Concepts relating tophysical impacts were mentioned by all patients

bFig. 1 BTC signs/symptoms by frequency and averagedisturbance rating (N = 23). For some interviews, insuf-ficient time was available to discuss all disturbance ratings.Consequently, the number of patients who mentioned adisturbance rating for a sign/symptom does not alwaysequal the number of patients who mentioned the sign/symptom. Eight signs/symptoms received no disturbancerating. Salient signs/symptoms are underlined. BTC biliarytract cancer

Emotional impacts

Physical impacts

Cognitive impacts

Inability to workFinancial burden Trouble meeting the

needs of the family

Impact on social lifeImpact on family

Difficulty with self-care(e.g. dressing, bathing)

Insomnia

Disruption/inconvenience in day-to-day activities Inability to travel

Salient concepts

Num

ber o

f pat

ient

s re

porti

ng im

pact

s

00

Not disturbing3 4 5 6 7 8 9 10

Very disturbing

5

10

15

20

25

Average disturbance rating

Fig. 2 BTC impacts by frequency and average disturbancerating (N = 23). For some interviews, insufficient time wasavailable to discuss all disturbance ratings. Consequently,the number of patients who mentioned a disturbancerating for an impact does not always equal the number of

patients who mentioned the impact. BTC biliary tractcancer

564 Oncol Ther (2021) 9:557–573

Table 2 Salient sign and symptom concepts and example quotations elicited from patients with BTC (N = 23)

Sign/symptom Patientsmentioning sign/symptom, n (%)a

Mean disturbancerating (number ofpatients rated)b

Example patient quotation (stage, subtype)

Fatigue/lack of

energy

23 (100) 7.9 (21) ‘‘I’m just fatigued. The fatigue is probably the biggest

[symptom] … Chemo obviously makes it a little

worse, but even just in general, like even between my

stints, I am just really tired.’’ (metastatic, perihilar)

Abdominal

pain

19 (83) 7.6 (16) ‘‘At the time of diagnosis, the [abdominal pain] was the

worst pain I’ve ever had. I remember thinking, ‘This

is worse than natural childbirth’. It was truly the

worst pain, I couldn’t move. I remember I was kind

of hunched over, and I couldn’t straighten up.’’

(metastatic, intrahepatic)

Lack of

appetite

18 (78) 5.8 (16) ‘‘… a lot of times, I don’t even have the sensation that

I’m hungry. I just have to eat knowing that you need

the nourishment and stuff. I’m still that way.’’

(locally advanced, intrahepatic)

Difficulty

eating/

feeling of

fullness

18 (78) 5.8 (13) ‘‘I would become hungry, but once I started to eat, I

became full very quickly.’’ (metastatic, intrahepatic)

Abdominal

bloating

16 (70) 5.8 (9) ‘‘Sometimes it’s like, ‘Gosh, I feel like I’m maybe too

full,’ and I kind of look down and it’s like, ‘You’re

looking a little puffy today.’ And then a day or so

later I don’t see it.’’ (locally advanced, intrahepatic)

Diarrhoea 17 (74) 6.1 (14) ‘‘That was pretty much throughout the whole six

months of the … the gastrointestinal issues … I

started having violent diarrhoea, and so I wasn’t

keeping any of the nutrients.’’ (locally advanced,

intrahepatic)

Nausea/

queasiness

15 (65) 6.4 (15) ‘‘… Nauseous all the time. I mainly just wanted to eat

simple foods, like soup or mashed potatoes.’’

(metastatic, intrahepatic)

Constipation 12 (52) 6.4 (10) ‘‘I can go from being extremely constipated to having

five, six, seven loose stools in a day.’’ (metastatic,

intrahepatic)

Insomnia 18 (78) 7.3 (9) ‘‘I live alone, and you get in bed, and your mind starts

wandering. Yeah. I have terrible problems with

insomnia.’’ (locally advanced, perihilar)

Oncol Ther (2021) 9:557–573 565

with the perihilar subtype and by 16 of 17patients with intrahepatic BTC. Financial bur-den was mentioned by nine of 17 patients withthe intrahepatic subtype and by two of fivepatients with the perihilar subtype. Across thedisease subtypes patients generally reportedsimilar salient signs/symptoms. Impacts werealso similar across subtypes. Physical impactswere mentioned by 16 of 17 patients withintrahepatic BTC and by five of five patientswith perihilar BTC. Emotional impacts werementioned by 13 of 17 patients with intrahep-atic BTC and by three of five patients with per-ihilar BTC. Cognitive impacts were mentioned

by 10 of 17 patients with intrahepatic BTC andby three of five patients with perihilar BTC. Thesingle patient with gall bladder BTC did notmention physical impacts but did mentionemotional and cognitive impacts.

Signs/Symptoms and Impacts by Disease StageWhile fatigue, insomnia and GI signs/symp-toms were salient signs/symptoms experiencedby patients at all disease stages, several signs/symptoms and impacts were particularly rele-vant in specific disease stages. A total of 27signs/symptoms and nine impacts were men-tioned by patients with early-stage BTC (n = 3).

Table 2 continued

Sign/symptom Patientsmentioning sign/symptom, n (%)a

Mean disturbancerating (number ofpatients rated)b

Example patient quotation (stage, subtype)

Other pain 14 (61) 6.9 (11) ‘‘To touch in my armpit areas hurts … it’s more just if

I touch it. Because I’m washing under my armpit or

if I were to just touch underneath my armpits, it feels

really tender and sore, which is a weird spot. Because

I’ve never had anything like it.’’ (early, intrahepatic)

Itchy skin 14 (61) 7.2 (13) ‘‘I would just have this terrible itching, mainly up in the

upper body but sometimes my legs too.’’ (early,

intrahepatic)

Fever/chills 15 (65) 6.4 (13) ‘‘It was horrible because I’d be hot and then take stuff

off and then I’d be cold, and I’d put it on. It was

back and forth like a yo-yo all day long.’’ (locally

advanced, gall bladder)

Sensitivity to

cold

13 (57) 6.1 (11) ‘‘Ever since I’ve been on chemotherapy, I find that I

don’t tolerate the cold like I used to. I get cold very

easily, and that’s something that has not gone away.’’

(locally advanced, intrahepatic)

Muscle loss 16 (70) 6.2 (11) ‘‘I think due to the fatigue and the lack of nourishment,

they [muscle] just atrophied.’’ (metastatic,

intrahepatic)

Patient-reported signs and symptoms of BTC could be related to disease and/or current or past treatmentsBTC biliary tract cancera Spontaneously mentioned or promptedb For some interviews, insufficient time was available to discuss all disturbance ratings. Consequently, the number ofpatients who mentioned a disturbance rating for a sign/symptom does not always equal the number of patients whomentioned the sign/symptom

566 Oncol Ther (2021) 9:557–573

Nail changes (which were not included in theall-stage conceptual model due to being deemedby patients and clinicians to be treatment-re-lated) and rash were signs/symptoms specific tothe early stage. Jaundice and weight loss werenot reported by patients with early-stage cancer.A total number of 41 signs/symptoms and 11impacts were mentioned by patients withlocally advanced BTC (n = 11). Three patientswith locally advanced BTC mentioned flu-likesymptoms (not including chills or cough),while rash, abdominal tenderness and intestinalgas were each only mentioned by one patient.The impact ‘‘inability to travel’’ was specific tolocally advanced BTC. For patients with meta-static BTC (n = 9), 41 signs/symptoms and 10impacts were mentioned. Signs/symptoms

specific to patients with metastatic BTC inclu-ded heartburn, problems with hearing, head-aches, intestinal pain, mouth sensitivity andhypercalcaemia.

Concept Saturation

Concept saturation was reached by the fourthand third of five interview waves for signs/symptoms and impacts, respectively. Of the 79concepts identified, 49% (39 concepts) werementioned in the first wave.

Table 3 Salient impact concepts and example quotations elicited from patients with BTC (N = 23)

Impact Patientsmentioningimpact, n (%)a

Mean disturbancerating (number ofpatients rated)b

Example patient quotation (stage, subtype)

Physical impacts

(difficulty

walking)

22 (96) 7.5 (15) ‘‘I’ve kind of, the last few months, just gotten to where

I ask for a wheelchair at the airport, because I

figure why should I spend all my energy trying to get

from one end of the airport to the other, when they

can just take me. I have given in to that symptom,

just because I have to work hard to breath when I’m

walking very far.’’ (metastatic, intrahepatic)

Emotional impacts

(depression)

17 (74) 8.5 (15) ‘‘Yeah, the depression, and it really, really hit me after

treatment, to the point to where my doctor added

an antidepressant for me.’’ (locally advanced,

intrahepatic)

Cognitive impacts

(memory loss,

fuzzy brain)

14 (61) 8.2 (13) ‘‘Or the brain fog and you can’t concentrate on

anything and [the children] want to tell me all about

the game they’re playing or how the day was. It’s like

it’s not sinking in at all. They could be speaking in a

foreign language and I wouldn’t have known.’’

(locally advanced, gall bladder)

BTC biliary tract cancera Spontaneously mentioned or promptedb For some interviews, insufficient time was available to discuss all disturbance ratings. Consequently, the number ofpatients who mentioned a disturbance rating for an impact does not always equal the number of patients who mentionedthe impact

Oncol Ther (2021) 9:557–573 567

Conceptual Model of the BTC PatientExperience

The final all-stage conceptual model for thepatient experience of BTC includes 50 signs/symptoms and 12 impacts (Fig. 3). Of these, 14signs/symptoms and three impacts were salient.Salient signs/symptoms were fatigue/lack ofenergy, abdominal pain, lack of appetite, diffi-culty eating/feeling of fullness, abdominalbloating, diarrhoea, nausea/queasiness, consti-pation, insomnia, other pain, itchy skin, fever/chills, sensitivity to cold and muscle loss. Allsalient impacts were reported in all three stagesof BTC and were physical impacts (e.g. difficultywalking), emotional impacts (e.g. depression)and cognitive impacts (e.g. memory loss, fuzzy

brain). An additional two signs/symptoms (hairloss and change in taste) were salient but werenot included in the final model because theywere deemed to be associated with some treat-ments (treatment-related signs/symptoms aresummarized in Supplementary Table 3).

DISCUSSION

Direct patient input through semi-structuredqualitative interviews provides a comprehensiveunderstanding of disease and treatment experi-ence from the patient perspective and is crucialto inform fit-for-purpose patient-centred out-come measurement strategies for clinical trials.In this qualitative study, targeted literature

Tiredness• Fatigue/lack of energy• Decreased energy/stamina

Pain/discomfort in abdominal area• Abdominal pain• Dull ache in abdominal

area• Intestinal pain• Able to feel tumour• Abdominal tenderness• Gall bladder pain

Appetite/eating• Lack of appetite• Difficulty eating/feeling

of fullness

Breathing• Shortness of breath• Difficulty breathing

Abdominal signs/symptoms• Abdominal bloating• Abdominal distress

GI signs/symptoms• Diarrhoea• Nausea/queasiness• Constipation• Vomiting• Change in stool (excluding

diarrhoeaand constipation)

• Heartburn

Insomnia• Insomnia

Pain and discomfort in other parts of the body• Head/ear pressure changes• Back pain• Other pain

Skin• Itchy skin• Jaundice/yellow skin/

yellow eyes• Rash

Extremities• Swollen ankles/arms/legs

General unwell feeling• Aching all over• General unwell feeling• Flu-like symptoms (not

necessarily including chills or cough)

• Fever/chills

Urogenital• Change in urine• Urinary tract infection• Urgency to urinate

Circulatory issues• Hypercalcaemia• Portal vein thrombosis• Blood clot• Hard veins

Senses• Sensitivity to cold• Sensitivity to noise

Strength• Muscle loss• Weakness• Muscle tension

Weight change• Losing weight• Weight gain

Mouth and throat signs/symptoms• Cough/throat irritation/voice

change• Difficulty swallowing• Cannot drink fluid• Mouth sensitivity

Signs/symptoms

• Physical impacts (e.g. difficulty walking)

• Emotional impacts (e.g. depression) • Cognitive impacts (e.g. memory loss

and fuzzy brain)

Additional impacts• Trouble meeting the needs

of the family• Impact on the family• Impact on social life

• Difficulty with self-care (e.g. dressing)

• Financial burden• Insomnia

• Inability to travel• Bother• Inability to work

Impacts

Fig. 3 All-stage disease-specific conceptual model for thepatient experience of BTC. Bold signifies salient concepts,defined as mentioned by[ 50% of patients (N = 23) and

an average disturbance rating of[ 5. BTC biliary tractcancer, GI gastrointestinal

568 Oncol Ther (2021) 9:557–573

searches and clinician interviews were con-ducted to develop a preliminary BTC concep-tual model and to guide patient interviews. Theconcept elicitation method was used to gain afurther understanding of the experience withBTC directly from patients. Following thepatient interviews, the preliminary conceptualmodel was refined and finalized to represent apatient-centred perspective of the lived experi-ence of BTC. To our knowledge, the all-stageconceptual model developed in the currentstudy is the first of its kind. Patients with BTCexperience reduced HRQoL [19, 36, 37], and thiswas reflected in the current study. Patients withBTC mentioned 46 signs/symptoms and eightimpacts of BTC that had not been identifiedfrom the targeted literature and clinician inter-views, highlighting the importance of obtaininginformation on the patient experience directlyfrom patients and suggesting a need to ensurefit-for-purpose patient-centred outcomes mea-surement in BTC clinical trials.

Signs/symptoms not previously identifiedfrom the targeted literature search and clinicianinterviews were related to pain and discomfort,GI signs/symptoms, senses, strength, weightgain, hair, mouth and throat signs/symptoms,skin, extremities and general unwell feeling.Several concepts that were not previouslyreported were variations of concepts identifiedin the targeted literature search and clinicianinterviews. For example, in the preliminary lit-erature search, insomnia was associated onlywith itching; however, patients interviewed inthis study mostly mentioned insomnia in asso-ciation with worries and anxieties, aching andpain, emphasizing the importance of perform-ing qualitative interviews in gaining patientperspectives.

Across disease stages and of all reportedconcepts, those rated most disturbing bypatients were emotional and cognitive impacts.Physical impacts, while more frequently repor-ted than emotional and cognitive impacts, weredeemed slightly less disturbing by patients.These findings demonstrate the significance ofpsychosocial assessment in BTC and suggest theneed for psychosocial supportive care forpatients with BTC at all stages of the disease.Overall, the all-stage disease-specific conceptual

model for BTC comprises 50 sign/symptomsand 12 impacts related to BTC. Seventeen ofthese concepts were deemed to be salient; thatis, they were prevalent and highly disturbing(reported by more than 50% of patients, with adisturbance rating of C 5 out of 10).

Diagnosis of BTC in patients typically occursat a later stage in the disease [8]; this is reflectedin this study by fewer patients recruited withearly-stage BTC than the number of patientswith locally advanced or metastatic stages. Thesmall sample size of patients with early-stageBTC (n = 3) in this study should be consideredwhen interpreting data, and suggests that thefindings of this study are more applicable todrug development in advanced disease. Patientswith early-stage BTC experienced a smallernumber of symptoms than did patients withlater stages of BTC, although some salient signs/symptoms that occurred at the early stage ofBTC were commonly reported throughout thedisease stages. The salient signs/symptomsexperienced by patients at all disease stageswere fatigue, insomnia and GI signs/symptoms.

This study had several strengths. It had aspecific focus on BTC, and the collection ofsigns/symptoms and impacts from qualitativepatient interviews was robust and overarching.Sample sizes were adequate, as demonstrated bythe saturation of concepts at the fourth andthird of five waves for signs/symptoms andimpacts, respectively. Further, the study popu-lation was representative in terms of age; apooling of data from several BTC studies calcu-lated a median age (interquartile range) of 58(51–65) years [38], which is comparable to thecurrent study in which a median age of 54 yearswas determined.

Various key factors influence the risk ofdeveloping BTC, including age, sex, ethnicityand the presence of comorbidities [38–40]. Thecohort of patients in this study was solely fromthe USA, which in general presents lower ratesof BTC than those seen in Asian and SouthAmerican populations [4]. Sex ratios can varyaccording to geographical location and ethnic-ity; for example, in Chile, of 23,716 deathsbetween 2000 and 2012 due to gall bladdercancer and extrahepatic cholangiocarcinomasubtypes, 74% were women [4]. Interestingly,

Oncol Ther (2021) 9:557–573 569

perihilar cholangiocarcinoma incidence andmortality in Thailand are generally higher formen than women [41]. In the USA (between1999 and 2013) and South Korea (between 2006and 2015), incidence rates for intrahepatic andextrahepatic cholangiocarcinoma were higherfor men than women [40, 42]. In the currentstudy, 78% of patients were women, which doesnot reflect the overall sex distribution andwhich may be due to the greater willingness ofwomen to participate in such studies comparedwith men. Most patients were college-educated.The sample size of patients with early-stage BTCwas small (n = 3). Most of the patients in thisstudy had intrahepatic cholangiocarcinoma(n = 17), which is not representative of thenatural distribution of BTC subtypes; gall blad-der cancer is the most common BTC subtypeworldwide [40]. Furthermore, no patients withextrahepatic distal BTC were recruited in thisstudy, which may be because the prevalence ofdistal BTC is much lower than that of perihilarBTC in the USA [43]. Responses are likely tohave been affected by treatments receivedaround the time of the interview. Based on thisstudy, recommendations for future researchmay include recruiting patients of differentethnicities and geographical locations andincreasing the sample sizes of patients withdifferent BTC subtypes to gain an even widerunderstanding of the patient experience of BTC.

CONCLUSIONS

In conclusion, patients with BTC reported arange of signs/symptoms and impacts of theirdisease that negatively affect their daily func-tioning and quality of life, including some thathad not been previously identified. The all-stageconceptual model developed in this study pro-vides insights into the experience of patientsliving with BTC, which can aid patient–clini-cian dialogue and act as a tool for informingpatients of what to expect at different diseasestages and possibly from their treatments.Clinical trials targeting BTC should assesspatient perspectives on how their disease andtreatment impacts their quality of life, to

provide a complementary view to traditionalefficacy and safety outcomes.

ACKNOWLEDGEMENTS

We thank the study participant(s) for theirinvolvement in the study.

Medical Writing/Editorial Assis-tance. Medical writing support was provided byOlivia McKenna, PhD, of Oxford PharmaGene-sis, Oxford, UK, and was funded by AstraZeneca.

Funding. This project was funded by Astra-Zeneca, Gaithersburg, MD, USA. Research andpublication fees are funded by the studysponsor.

Authorship. All named authors meet theInternational Committee of Medical JournalEditors (ICMJE) criteria for authorship for thisarticle, take responsibility for the integrity ofthe work as a whole, and have given theirapproval for this version to be published.

Authorship Contributions. Design andconception: Nikunj Patel, Nana Rokutanda,Christina Workman, Oren Meyers, GordonCohen, Xandra Lie, Chad Gwaltney; First draft:Nikunj Patel, Nana Rokutanda, ChristinaWorkman, Oren Meyers, Gordon Cohen, Xan-dra Lie, Chad Gwaltney, Afsaneh Barzi, DavideMelisi, Teresa Macarulla, Makoto Ueno,Melinda Bachini. All authors revised criticallyand approved the final version for publication.

Disclosures. Nikunj Patel, Nana Rokutanda,Christina Workman and Gordon Cohen areemployees of AstraZeneca and hold shares inAstraZeneca. Xandra Lie and Oren Myers areemployees of IQVIA, which received funds fromAstraZeneca to conduct the analysis of the studydata. Chad Gwaltney received consulting feesfrom IQVIA to participate in the design, exe-cution and analysis of this study. MelindaBachini has received honorarium from Taiho,Quod Erat Demonstrandum (QED) clinical ser-vices Ltd, Incyte and EMD Serono (for partici-pating in patient advisory boards, review of

570 Oncol Ther (2021) 9:557–573

patient materials and speaking engagements).Afsaneh Barzi, Davide Melisi, Teresa Macarulla,Makoto Ueno and Seung Tae Kim have nothingto declare.

Ethics Approval. The qualitative interview,research protocol, interview guide, and allpatient communication documents werereviewed and approved by the New EnglandInstitutional Review Board (NEIRB) in January2019.

Data Availability. The datasets generatedduring and/or analysed during the currentstudy are available from the correspondingauthor on reasonable request.

Consent to Participate. All patients pro-vided consent to participate online.

Consent for Publication. All participatingpatients provided consent online.

Open Access. This article is licensed under aCreative Commons Attribution-NonCommer-cial 4.0 International License, which permitsany non-commercial use, sharing, adaptation,distribution and reproduction in any mediumor format, as long as you give appropriate creditto the original author(s) and the source, providea link to the Creative Commons licence, andindicate if changes were made. The images orother third party material in this article areincluded in the article’s Creative Commonslicence, unless indicated otherwise in a creditline to the material. If material is not includedin the article’s Creative Commons licence andyour intended use is not permitted by statutoryregulation or exceeds the permitted use, youwill need to obtain permission directly from thecopyright holder. To view a copy of this licence,visit http://creativecommons.org/licenses/by-nc/4.0/.

REFERENCES

1. Tariq NU, McNamara MG, Valle JW. Biliary tractcancers: current knowledge, clinical candidates and

future challenges. Cancer Manag Res. 2019;11:2623–42.

2. Taylor-Robinson SD, Toledano MB, Arora S, KeeganTJ, Hargreaves S, Beck A, et al. Increase in mortalityrates from intrahepatic cholangiocarcinoma inEngland and Wales 1968–1998. Gut. 2001;48(6):816–20.

3. Saha SK, Zhu AX, Fuchs CS, Brooks GA. Forty-yeartrends in cholangiocarcinoma incidence in the US:intrahepatic disease on the rise. Oncologist.2016;21(5):594–9.

4. Marcano-Bonilla L, Mohamed EA, Mounajjed T,Roberts LR. Biliary tract cancers: epidemiology,molecular pathogenesis and genetic risk associa-tions. Chin Clin Oncol. 2016;5(5):61.

5. Stinton LM, Shaffer EA. Epidemiology of gallblad-der disease: cholelithiasis and cancer. Gut Liver.2012;6(2):172–87.

6. Bridgewater J, Galle PR, Khan SA, Llovet JM, ParkJW, Patel T, et al. Guidelines for the diagnosis andmanagement of intrahepatic cholangiocarcinoma.J Hepatol. 2014;60(6):1268–89.

7. Bridgewater JA, Goodman KA, Kalyan A, MulcahyMF. Biliary tract cancer: epidemiology, radiother-apy, and molecular profiling. Am Soc Clin OncolEduc Book. 2016;35:e194–203.

8. Lamarca A, Edeline J, McNamara MG, Hubner RA,Nagino M, Bridgewater J, et al. Current standardsand future perspectives in adjuvant treatment forbiliary tract cancers. Cancer Treat Rev. 2020;84:101936.

9. Anderson C, Kim R. Adjuvant therapy for resectedextrahepatic cholangiocarcinoma: a review of theliterature and future directions. Cancer Treat Rev.2009;35(4):322–7.

10. DeOliveira ML, Cunningham SC, Cameron JL,Kamangar F, Winter JM, Lillemoe KD, et al.Cholangiocarcinoma: thirty-one-year experiencewith 564 patients at a single institution. Ann Surg.2007;245(5):755–62.

11. Endo I, Gonen M, Yopp AC, Dalal KM, Zhou Q,Klimstra D, et al. Intrahepatic cholangiocarcinoma:rising frequency, improved survival, and determi-nants of outcome after resection. Ann Surg.2008;248(1):84–96.

12. Rizvi S, Khan SA, Hallemeier CL, Kelley RK, GoresGJ. Cholangiocarcinoma—evolving concepts andtherapeutic strategies. Nat Rev Clin Oncol.2018;15(2):95–111.

Oncol Ther (2021) 9:557–573 571

13. Butt Z, Parikh ND, Beaumont JL, Rosenbloom SK,Syrjala KL, Abernethy AP, et al. Development andvalidation of a symptom index for advanced hepa-tobiliary and pancreatic cancers: the NationalComprehensive Cancer Network FunctionalAssessment of Cancer Therapy (NCCN-FACT)Hepatobiliary-Pancreatic Symptom Index (NFHSI).Cancer. 2012;118(23):5997–6004.

14. Woradet S, Promthet S, Songserm N, Parkin DM.Factors affecting health-related quality of life inpatients with cholangiocarcinoma in the north-eastern region of Thailand. Cancer Nurs.2015;38(6):E46–51.

15. Woradet S, Songserm N, Promthet S, Parkin DM.Health-related quality of life and survival ofcholangiocarcinoma patients in northeasternregion of Thailand. PLoS ONE. 2016;11(9):e0163448.

16. Murad SD, Heimbach JK, Gores GJ, Rosen CB,Benson JT, Kim WR. Excellent quality of life afterliver transplantation for patients with perihilarcholangiocarcinoma who have undergone neoad-juvant chemoradiation. Liver Transpl. 2013;19(5):521–8.

17. Kaupp-Roberts SD, Yadegarfar G, Friend E, O’Don-nell CM, Valle JW, Byrne C, et al. Validation of theEORTC QLQ-BIL21 questionnaire for measuringquality of life in patients with cholangiocarcinomaand cancer of the gallbladder. Br J Cancer.2016;115(9):1032–8.

18. Dengsø K, Bangsgaard L, Marcussen AM, Chris-tensen BM, Hillingsø J, Thomsen T. Patients’descriptions of initial symptoms of cholangicarci-noma and the journey to diagnosis: a qualitativestudy. J Neoplasm. 2017;2(2):14.

19. Friend E, Yadegarfar G, Byrne C, Johnson CD, SezerO, Pucciarelli S, et al. Development of a question-naire (EORTC module) to measure quality of life inpatients with cholangiocarcinoma and gallbladdercancer, the EORTC QLQ-BIL21. Br J Cancer.2011;104(4):587–92.

20. Somjaivong B, Thanasilp S, Preechawong S, Sloan R.The influence of symptoms, social support, uncer-tainty, and coping on health-related quality of lifeamong cholangiocarcinoma patients in northeastThailand. Cancer Nurs. 2011;34(6):434–42.

21. Brod M, Tesler LE, Christensen TL. Qualitativeresearch and content validity: developing bestpractices based on science and experience. Qual LifeRes. 2009;18(9):1263–78.

22. CHMP. Reflection paper on the regulatory guidancefor the use of health-related quality of life (HRQL)

measures in the evaluation of medicinal products.London: European Medicines Agency; 2005.

23. Food and Drug Administration. Guidance forindustry: patient-reported outcome measures: usein medical product development to support label-ing claims. Fed Regist. 2009;74(235):65132–3.

24. Patrick DL, Burke LB, Gwaltney CJ, Leidy NK,Martin ML, Molsen E, et al. Content validity—establishing and reporting the evidence in newlydeveloped patient-reported outcomes (PRO)instruments for medical product evaluation: ISPORPRO good research practices task force report: part1—eliciting concepts for a new PRO instrument.Value Health. 2011;14(8):967–77.

25. Koeberle D, Saletti P, Borner M, Gerber D, DietrichD, Caspar CB, et al. Patient-reported outcomes ofpatients with advanced biliary tract cancers receiv-ing gemcitabine plus capecitabine: a multicenter,phase II trial of the Swiss Group for Clinical CancerResearch. J Clin Oncol. 2008;26(22):3702–8.

26. Mikkelsen MK, Nielsen DL, Vinther A, Lund CM,Jarden M. Attitudes towards physical activity andexercise in older patients with advanced cancerduring oncological treatment—a qualitative inter-view study. Eur J Oncol Nurs. 2019;41:16–23.

27. Ibrahim F, Sandstrom P, Bjornsson B, Larsson AL,Drott J. ‘‘I want to know why and need to beinvolved in my own care…’’: a qualitative interviewstudy with liver, bile duct or pancreatic cancerpatients about their experiences with involvementin care. Support Care Cancer. 2019;27(7):2561–7.

28. Krippendorff K. Content analysis: an introductionto its methodology. Beverly Hills: Sage Publications;1980.

29. Fusch P, Ness LR. Are we there yet? Data saturationin qualitative research. Qual Rep. 2015;20(9):1408–16.

30. Guest G, Bunce A, Johnson L. Howmany interviewsare enough?: An experiment with data saturationand variability. Field Methods. 2006;18(1):59–82.

31. Turner-Bowker DM, Lamoureux RE, Stokes J,Litcher-Kelly L, Galipeau N, Yaworsky A, et al.Informing a priori sample size estimation in quali-tative concept elicitation interview studies forclinical outcome assessment instrument develop-ment. Value Health. 2018;21(7):839–42.

32. Elberg Dengso K, Hillingso J, Marcussen AM,Thomsen T. Health-related quality of life and anx-iety and depression in patients diagnosed withcholangiocarcinoma: a prospective cohort study.Acta Oncol. 2017;56(2):198–204.

572 Oncol Ther (2021) 9:557–573

33. Shun SC, Chiou JF, Lai YH, Yu PJ, Wei LL, Tsai JT,et al. Changes in quality of life and its related fac-tors in liver cancer patients receiving stereotacticradiation therapy. Support Care Cancer. 2008;16(9):1059–65.

34. Wu TT, Li WM, Li HC, Ao GK, Zheng F, Lin H.Percutaneous intraductal radiofrequency ablationfor extrahepatic distal cholangiocarcinoma: amethod for prolonging stent patency and achievingbetter functional status and quality of life. Cardio-vasc Intervent Radiol. 2017;40(2):260–9.

35. Koyama K, Fukunishi I, Kudo M, Sugawara Y,Makuuchi M. Psychiatric symptoms after hepaticresection. Psychosomatics. 2003;44(1):86–7.

36. Yang Z, Wan C, Li W, Cun Y, Meng Q, Ding Y, et al.Development and validation of the simplified Chi-nese version of EORTC QLQ-HCC18 for patientswith hepatocellular carcinoma. Cancer Invest.2015;33(8):340–6.

37. Bridgewater J, Lopes A, Palmer D, Cunningham D,Anthoney A, Maraveyas A, et al. Quality of life,long-term survivors and long-term outcome fromthe ABC-02 study. Br J Cancer. 2016;114(9):965–71.

38. McGee EE, Jackson SS, Petrick JL, Van Dyke AL,Adami HO, Albanes D, et al. Smoking, alcohol, and

biliary tract cancer risk: a pooling project of 26prospective studies. J Natl Cancer Inst.2019;111(12):1263–78.

39. Castro FA, Koshiol J, Hsing AW, Devesa SS. Biliarytract cancer incidence in the United States—de-mographic and temporal variations by anatomicsite. Int J Cancer. 2013;133(7):1664–71.

40. Van Dyke AL, Shiels MS, Jones GS, Pfeiffer RM,Petrick JL, Beebe-Dimmer JL, et al. Biliary tractcancer incidence and trends in the United States bydemographic group, 1999–2013. Cancer.2019;125(9):1489–98.

41. Sripa B, Pairojkul C. Cholangiocarcinoma: lessonsfrom Thailand. Curr Opin Gastroenterol.2008;24(3):349–56.

42. Kim BW, Oh CM, Choi HY, Park JW, Cho H, Ki M.Incidence and overall survival of biliary tract can-cers in South Korea from 2006 to 2015: using theNational Health Information Database. Gut Liver.2019;13(1):104–13.

43. Banales JM, Marin JJG, Lamarca A, Rodrigues PM,Khan SA, Roberts LR, et al. Cholangiocarcinoma2020: the next horizon in mechanisms and man-agement. Nat Rev Gastroenterol Hepatol.2020;17(9):557–88.

Oncol Ther (2021) 9:557–573 573