Abstract
Background
Biliary tract cancers (BTCs), a heterogeneous group of cancers arising from the biliary tract, account for ∼15% of primary liver cancers. Patients are often diagnosed with advanced disease, resulting in poor prognoses and documented 5-year survival rates of <2%. The aim of this study was to describe real-world treatment patterns and outcomes among patients with advanced BTC in France, Germany, Italy, Spain, and the UK.
Patients and methods
A retrospective, physician-abstracted chart review survey was conducted between May 2018 and October 2021. Data were abstracted from medical charts of adult patients diagnosed with advanced BTC who initiated treatment with first-line (1L) systemic therapy. Patient characteristics, treatment patterns, and clinical outcomes were summarized descriptively.
Results
In total, 196 physicians provided data for 792 advanced BTC patients who initiated 1L systemic therapy. Mean age was 65.7 years and most of the patients were male (62.6%). At data abstraction 56.1% were deceased. The most frequently prescribed 1L systemic treatment was cisplatin–gemcitabine (47.9%). The median 1L treatment duration was 5.3 months. Additionally, 33.5% of patients received 2L and 4.5% received 3L+ treatment. Median real-world overall survival from index date was 13.4 months, with substantial regional variation.
Conclusions
This study supports extant data detailing that presentation at an advanced stage may contribute to the poor outcomes of BTC patients. There are significant implications for awareness and guideline-driven pathway changes to improve outcomes for these poorly served patients.
Key words: biliary tract cancer, physician-based chart review survey, Europe, treatment effectiveness, treatment patterns
Highlights
-
•
Real-world treatment patterns and outcomes were investigated in advanced BTC.
-
•
The GARNET-2 chart review was conducted in France, Germany, Italy, Spain, and the UK.
-
•
The most common 1L systemic treatment was cisplatin–gemcitabine (47.9%).
-
•
Median real-world overall survival was 13.4 months with regional variation.
-
•
Findings underscore the need for new advanced BTC treatments to improve survival.
Introduction
Biliary tract cancers (BTCs) comprise a diverse group of malignancies, encompassing cholangiocarcinoma (CCA), gallbladder cancers (GBC), and ampullary cancers.1,2 Among CCAs, perihilar CCA (pCCA), subtype of extrahepatic CCA (eCCA), is the most common subtype, accounting for ∼60%, followed by distal CCA (dCCA; ∼30%) and intrahepatic CCA (iCCA; <10%).3 The published incidence and mortality of CCA in Europe ranges from 0.5 to 3.4 cases per 100 000 and 1.6 to 5.7 per 100 000, respectively.4,5 Patients with eCCA tend to present with obstructive jaundice in contrast to iCCA patients who experience non-specific symptoms, such as general malaise and abdominal pain, frequently leading to late diagnosis with more advanced disease. Prognoses for these patients are poor, and 5-year survival rates are <2% for metastatic BTC.6
There has been misclassification and misdiagnosis of BTCs,3,7,8 reflected in ICD-11 coding, currently being introduced. This may have contributed to the apparent increasing incidence of iCCA in western Europe together with increased awareness of the diagnosis and the establishment of standard of care therapy treatment.4 Radical surgical resection is the only curative treatment of BTC, but only about 10%-35% of cases are considered resectable.9
Understanding the treatment landscape and survival rates for patients diagnosed with advanced BTC in the real-world setting may provide insights on the clinical practice in Europe. The aim of this study was to describe the real-world treatment patterns, clinical outcomes, and health care resource use (HCRU) among BTC patients who initiated first-line (1L) systemic therapy in France, Germany, Italy, Spain, and the UK.
Methods
Study design
This was a non-interventional, retrospective, physician-abstracted, web-based chart review of patients with a recent confirmed diagnosis of advanced BTC who underwent treatment. The protocol and data abstraction form were reviewed, and the study was given exemption from additional oversight by the Advarra institutional review board (IRB). Data collection was fully anonymized and conducted according to the European Society of Marketing and Opinion Research (ESOMAR), European Pharmaceutical Market Research Association (EphMRA), and British Healthcare Business Intelligence Association (BHBIA) rules and guidelines to ensure ethical conduct under health care market research regulation. Physicians based in France, Germany, Italy, Spain, and the UK were recruited from a database of specialists maintained by data vendor A + A Healthcare. Participating physicians were required to be certified in medical oncology, hepatology, gastroenterology, or in internal medicine, have at least 3 years of experience treating BTC patients, and have personally treated or managed at least two patients (per month). Physicians provided informed consent to participate in the study. In order to ensure that patient charts were selected randomly, the survey generated a random letter from the alphabet. Physicians were then asked to select an eligible patient whose initial matched with the randomly generated letter. The medical charts of selected patients were then reviewed, and anonymized patient-level information on clinical characteristics, treatment, and outcomes were abstracted. No participant-identifying data were collected during the data abstraction process, or at any time thereafter.
Patient population
Each physician reported anonymized patient-level information for up to six patients who met the selection criteria. Adults (age ≥18) diagnosed with advanced BTC between May 2018 and October 2021 who initiated 1L systemic treatment (Figure 1) were included in the study. The index date was defined as the initiation of 1L systemic therapy on or after the diagnosis date for advanced BTC. The follow-up period was defined from the index date until the date of the patient’s last recorded medical appointment or in-person treatment. The study required the availability of patient records for at least 8 months from the date of their diagnosis to ensure completeness of records in the sample. Patients who died during this period, however, were included in the study. Patients were excluded from the study if they had participated in a clinical trial, received 1L durvalumab as part of treatment of BTC (because the study period was before the 2022 approval of immunotherapy treatment of BTC in Europe), or had evidence of other prior or concurrent malignancy (except nonmelanoma skin cancer) at the time of data extraction.
Figure 1.
Study design schema.
BTC, biliary tract cancer; Oct, October; Q4, fourth quarter. ∗Patients were included in the study if they had a minimum post-diagnosis follow-up period of 8 months unless they died earlier.
Data collection
Physicians used an electronic data abstraction form (eDAF) to abstract data from the patients’ medical charts (see Supplementary File S1, available at https://doi.org/10.1016/j.esmorw.2024.100060). The data collected included demographic and clinical characteristics at diagnosis, total follow-up duration and BTC-related HCRU from diagnosis to the end of the observation period, and the time to treatment initiation from diagnosis. Additionally, data on treatment patterns (including treatment duration, discontinuation, and time to next treatment) and treatment outcomes (including treatment response, disease progression, and survival) were captured from treatment initiation to the end of the observation period.
Calculated outcomes were defined as follows:
-
•
Real-world time to next treatment (rwTTNT): time from discontinuation of one line of treatment to initiation of the subsequent treatment line.
-
•
Real-world overall survival (rwOS): from initiation of the 1L treatment to the date of death by any cause.
-
•
Annualized HCRU rates: considered all events occurring after diagnosis, accounting for the length of follow-up.
Analytic methods
Descriptive summary statistics including mean, standard deviation, median, and interquartile range (IQR) for continuous variables and frequency counts and percentages for categorical variables were generated for all study variables. Time to event analyses were estimated using the Kaplan–Meier (KM) method. Estimates of the median time to event, as well as rates at 1 year were evaluated. Confidence intervals (CIs) were calculated using Greenwood standard error. Rates and 95% CIs for HCRU were evaluated using negative binomial models (or Poisson models in case of estimation issues due to failed convergence) including an offset of patient follow-up time. A Sankey alluvial flow diagram was used to present treatment patterns reflecting systemic treatment transitions (1L to 2L, 2L to 3L). All results were reported for overall population and stratified by country. Statistical analyses were carried out using SAS Software, Version 9.4 or higher (SAS Institute, Cary, NC).
Results
Demographics and clinical characteristics
A total of 196 physicians (39 from France, 47 from Germany, 36 from Italy, 41 from Spain, and 33 from the UK), participated in the study. The most common physician specialty was medical oncologist (70.4%), followed by hepatologist (9.2%) and gastroenterologists (6.1%; see Supplementary Table S1, available at https://doi.org/10.1016/j.esmorw.2024.100060).
Data for 792 patients (167 France, 150 Germany, 153 Italy, 167 from Spain, and 155 UK) were included in the study. Patient characteristics in the overall sample and across the five countries are shown in Table 1, Supplementary Table S2A and B, available at https://doi.org/10.1016/j.esmorw.2024.100060, respectively. The estimated mean age was 65.7 years, and 62.6% of the patients were male. More than half of the patients were deceased at the time of data abstraction (n = 444; 56.1%). At the time of advanced BTC diagnosis, the majority of patients had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1 (79.7%), ranging from 14.7% in Germany to 32.9% in the UK. The percentage of patients with ECOG performance status of 2+ ranged from 8.5% in Italy to 22.0% and 22.2% in Germany and France, respectively. The most commonly reported risk factors at the time of advanced BTC diagnosis were alcohol use (34.8%), tobacco use (20.7%), and bile duct stones (20.3%), with gallstones, obesity, and type 2 diabetes observed in almost 20% of the patients.
Table 1.
Patient demographics and clinical characteristics
| Patients’ characteristics | Overall (N = 792) |
|---|---|
| Sex, n (%) | |
| Male | 496 (62.6) |
| Female | 296 (37.4) |
| Race, n (%) | |
| White | 711 (89.8) |
| Middle Eastern | 26 (3.3) |
| African/Black | 23 (2.9) |
| Other race group | 27 (3.4) |
| Not available in the patient’s medical chart | 5 (0.6) |
| Age (years) on the date of advanced BTC diagnosis, n (%) | |
| 18-44 | 27 (3.4) |
| 45-64 | 308 (38.9) |
| 65-74 | 335 (42.3) |
| ≥75 | 122 (15.4) |
| Site of primary BTC tumor at BTC diagnosis, n (%) | |
| Ampullary cancer/ampulla of Vater cancer | 81 (10.2) |
| Extrahepatic cholangiocarcinoma (eCCA) | 237 (29.9) |
| Distal (dCCA)a | 105 (44.3) |
| Perihilar/hilar (pCCA)a | 111 (46.8) |
| Subtype unknowna | 21 (8.9) |
| Gallbladder cancer (GBC) | 186 (23.5) |
| Intrahepatic cholangiocarcinoma (iCCA) | 276 (34.8) |
| Not available in the patient’s medical chart | 12 (1.5) |
| Genetic mutations/fusions detected after BTC diagnosisb, n (%) | |
| Epithelial growth factor receptor (EGFR) | 92 (11.6) |
| Fibroblast growth factor receptor 2 (FGFR2) | 87 (11.0) |
| Serine/threonine-protein kinase B-Raf (BRAF) | 61 (7.7) |
| Vascular endothelial growth factor (VEGF) | 51 (6.4) |
| Isocitrate dehydrogenase NADP+ 1 (IDH1) | 38 (4.8) |
| Receptor tyrosine-protein kinase erbB-2 (HER2) | 43 (5.4) |
| Other | 25 (3.2) |
| BTC risk factors and other attributable at BTC diagnosisc, n (%) | |
| Alcohol use | 276 (34.8) |
| Tobacco use | 164 (20.7) |
| Bile duct stones | 161 (20.3) |
| Gallstones | 156 (19.7) |
| Obesity | 153 (19.3) |
| Type 2 diabetes | 150 (18.9) |
| Cirrhosis | 130 (16.4) |
| Non-alcoholic fatty liver disease (NAFLD) | 52 (6.6) |
| Non-alcoholic steatohepatitis (NASH) | 44 (5.6) |
| Hepatitis B virus (HBV) infection | 39 (4.9) |
| Chronic biliary tract infection | 36 (4.5) |
| Hepatitis C virus (HCV) infection | 30 (3.8) |
| HCV with cirrhosisd | 10 (33.3) |
| Choledochal cyst | 24 (3.0) |
| Primary sclerosing cholangitis (PSC) | 22 (2.8) |
| Type 1 diabetes | 22 (2.8) |
| Drug-induced liver injury | 9 (1.1) |
| Ulcerative colitis | 9 (1.1) |
| Liver fluke | 7 (0.9) |
| Other | 24 (3.0) |
| Not available in the patient’s medical chart | 76 (9.6) |
| ECOG performance status at advanced BTC diagnosis, n (%) | |
| 0 | 180 (22.7) |
| 1 | 451 (56.9) |
| 2+ | 120 (15.2) |
| Not available in the patient’s medical chart | 41 (5.2) |
| Presence/absence of jaundice at advanced BTC diagnosis, n (%) | |
| Yes | 414 (52.3) |
| Received stent, n (%) | 254 (61.4) |
| No | 336 (42.4) |
| Not available in the patient’s medical chart, n (%) | 42 (5.3) |
| Presence/absence of biliary drainage at advanced BTC diagnosisc, n (%) | |
| Yes—external drainage | 93 (11.7) |
| Yes—stent | 305 (38.5) |
| Plastic stent | 104 (34.1) |
| Metallic stent | 195 (63.9) |
| Both plastic and metallic stent | 6 (2.0) |
| No drainage/stent | 366 (46.2) |
| Not available in the patient’s medical chart | 39 (4.9) |
| Treatment before advanced BTC diagnosisc, n (%) | |
| No prior treatment | 585 (73.9) |
| Biliary reconstruction | 24 (3.0) |
| Surgical resection | 111 (14.0) |
| Radiation therapy | 24 (3.0) |
| Other | 10 (1.3) |
| Unknown | 46 (5.8) |
AC, ampullary cancer; BTC, biliary tract cancer; dCCA, distal cholangiocarcinoma; eCC, extrahepatic cholangiocarcinoma; ECOG PS, Eastern Cooperative Oncology Group performance status; iCCA, intrahepatic cholangiocarcinoma; pCCA, perihilar cholangiocarcinoma.
Percentages are based on the number of patients with an extrahepatic cholangiocarcinoma.
Percentages are calculated on the overall sample. Patients without a test ranged from approximately 30% to 90% depending on the mutation type and the country. Additional mutation information was not collected.
Percentages may add up to more than 100% as patients may be counted in more than one category.
Percentages are based on the number of patients with HCV.
Patients who received only one line of therapy compared with those who received two or more lines of therapy had numerically a higher percentage of older patients (age 60+ years: 78.6% versus 66.0%) and a higher proportion of deaths (61.9% versus 44.5%) at date of abstraction than patients who received two or more lines of therapy.
Treatment patterns and HCRU
Among the 792 patients, the median follow-up duration from diagnosis was 13.5 months. The median (IQR) time from BTC diagnosis to initiation of 1L systemic treatment was 4.1 weeks (2.1-7.2 weeks), with no noticeable differences across countries. Of the 792 patients, 78 (9.8%) waited more than 13 weeks (3 months) from advanced BTC diagnosis to initiation of 1L systemic treatment. The most common reasons for delay were patient decision (n = 23; 40.4%), awaiting test results (n = 15; 26.3%), and need for a procedure or surgery (n = 9; 15.8%).
During the follow-up period, 527 (66.9%) received only one line of systemic treatment, 229 (29.1%) received two lines, 22 (2.8%) received three lines, and 10 (1.3%) received four or more lines (Table 2). Across the overall patient population, the most commonly used chemotherapy regimen was cisplatin–gemcitabine received by almost half of patients (47.9%). The percentage receiving cisplatin–gemcitabine 1L among those with a baseline ECOG score of 0 or 1 was 57.2% overall and ranged from 35.5% in France up to 78.8% in the UK.
Table 2.
Systemic treatment regimens
| Overall (N = 792) | France (N = 167) | Germany (N = 150) | Italy (N = 153) | Spain (N = 167) | UK (N = 155) | |
|---|---|---|---|---|---|---|
| Number of systemic treatment lines for advanced BTC, n (%) | ||||||
| 1 | 527 (66.9) | 125 (74.9) | 92 (63.0) | 90 (58.8) | 109 (65.3) | 111 (71.6) |
| 2 | 229 (29.1) | 34 (20.4) | 42 (28.8) | 56 (36.6) | 56 (33.5) | 41 (26.5) |
| 3 | 22 (2.8) | 5 (3.0) | 6 (4.1) | 7 (4.6) | 2 (1.2) | 2 (1.3) |
| 4 or more | 10 (1.3) | 3 (1.8) | 6 (4.1) | 0 | 0 | 1 (0.6) |
| Time from advanced BTC diagnosis to systemic treatment initiation, weeks | ||||||
| Median (IQR) | 4.1 (2.1-7.2) | 4.6 (2.3-8.0) | 4.1 (2.1-6.3) | 4.3 (2.6-7.0) | 3.4 (2.0-6.3) | 4.1 (2.3-8.3) |
| Patients with rwTTI ≥13 weeks (3 months), n (%) | 78 (9.8) | 21 (12.6) | 14 (9.3) | 7 (4.6) | 18 (10.8) | 18 (11.6) |
| Patients with rwTTI ≥26 weeks (6 months), n (%) | 42 (5.3) | 12 (7.2) | 9 (6.0) | 3 (2.0) | 12 (7.2) | 6 (3.9) |
| 1L systemic treatment regimen, n (%) | ||||||
| Cisplatin + gemcitabine | 379 (47.9) | 48 (28.7) | 45 (30.0) | 107 (69.9) | 74 (44.3) | 105 (67.7) |
| Gemcitabine | 67 (8.5) | 14 (8.4) | 11 (7.3) | 18 (11.8) | 14 (8.4) | 10 (6.5) |
| 5-Fluorouracil | 38 (4.8) | 13 (7.8) | 8 (5.3) | 0 | 3 (1.8) | 14 (9.0) |
| Gemcitabine + oxaliplatin | 37 (4.7) | 22 (13.2) | 5 (3.3) | 4 (2.6) | 6 (3.6) | 0 |
| Carboplatin + gemcitabine | 36 (4.5) | 10 (6.0) | 3 (2.0) | 1 (0.7) | 13 (7.8) | 9 (5.8) |
| Capecitabine | 30 (3.8) | 6 (3.6) | 2 (1.3) | 15 (9.8) | 5 (3.0) | 2 (1.3) |
| 5-Fluorouracil + cisplatin | 26 (3.3) | 6 (3.6) | 6 (4.0) | 1 (0.7) | 12 (7.2) | 1 (0.6) |
| Carboplatin | 11 (1.4) | 3 (1.8) | 4 (2.7) | 0 | 3 (1.8) | 1 (0.6) |
| Capecitabine + gemcitabine | 10 (1.3) | 3 (1.8) | 0 | 0 | 6 (3.6) | 1 (0.6) |
| Other combination | 100 (12.6) | 29 (17.4) | 47 (31.3) | 3 (2.0) | 16 (9.6) | 5 (3.2) |
| Other single agent | 15 (1.9) | 4 (2.4) | 9 (6.0) | 0 | 1 (0.6) | 1 (0.6) |
| Reason(s) for selecting 1L systemic treatment, n (%)a | ||||||
| Clinical evidence (e.g. improved survival) | 578 (73.0) | 118 (70.7) | 78 (52.0) | 124 (81.0) | 131 (78.4) | 127 (81.9) |
| ECOG performance status | 356 (44.9) | 80 (47.9) | 38 (25.3) | 80 (52.3) | 89 (53.3) | 69 (44.5) |
| Safety/tolerability profile of the chosen treatment | 122 (15.4) | 25 (15.0) | 26 (17.3) | 18 (11.8) | 37 (22.2) | 16 (10.3) |
| Patient preference | 82 (10.4) | 12 (7.2) | 33 (22.0) | 7 (4.6) | 12 (7.2) | 18 (11.6) |
| Patient demographics (e.g. age) | 75 (9.5) | 17 (10.2) | 22 (14.7) | 9 (5.9) | 12 (7.2) | 15 (9.7) |
| Presence of comorbidities (e.g. hypertension, poor renal function) | 65 (8.2) | 17 (10.2) | 11 (7.3) | 7 (4.6) | 23 (13.8) | 7 (4.5) |
| Treatment cost/reimbursement | 62 (7.8) | 17 (10.2) | 10 (6.7) | 12 (7.8) | 14 (8.4) | 9 (5.8) |
| Reason for discontinuation of 1L systemic treatment, n (%) | ||||||
| Disease progression | 356 (49.2) | 62 (46.3) | 57 (41.3) | 87 (58.4) | 94 (59.5) | 56 (38.6) |
| Death | 51 (7.0) | 15 (11.2) | 11 (8.0) | 1 (0.7) | 9 (5.7) | 15 (10.3) |
| Toxicity or adverse effect | 42 (5.8) | 17 (12.7) | 11 (8.0) | 6 (4.0) | 3 (1.9) | 5 (3.4) |
| Patient decision | 42 (5.8) | 10 (7.5) | 15 (10.9) | 5 (3.4) | 6 (3.8) | 6 (4.1) |
| Lost to follow-up | 18 (2.5) | 7 (5.2) | 8 (5.8) | 1 (0.7) | 0 | 2 (1.4) |
| Completed treatment course as planned | 180 (24.9) | 15 (11.2) | 26 (18.8) | 46 (30.9) | 41 (25.9) | 52 (35.9) |
| Lack of efficacy | 27 (3.7) | 5 (3.7) | 7 (5.1) | 3 (2.0) | 3 (1.9) | 9 (6.2) |
| Other | 8 (1.1) | 3 (2.2) | 3 (2.2) | 0 | 2 (1.3) | 0 |
| 2L systemic treatment regimen, n (%) | d | |||||
| None | 527 (66.5) | 125 (74.9) | 92 (61.3) | 90 (58.8) | 109 (65.3) | 111 (71.6) |
| Any | 265 (33.5) | 42 (25.1) | 58 (38.7) | 63 (41.2) | 58 (34.7) | 44 (28.4) |
| 5-Fluorocuracil + oxaliplatin | 57 (21.5) | 4 (9.5) | 3 (5.2) | 24 (38.1) | 18 (31.0) | 8 (18.2) |
| Capecitabine | 44 (16.6) | 4 (9.5) | 6 (10.3) | 5 (7.9) | 22 (37.9) | 7 (15.9) |
| 5-Fluorouracil + folinic acid + oxaliplatin | 21 (7.9) | 2 (4.8) | 3 (5.2) | 5 (7.9) | 1 (1.7) | 10 (22.7) |
| 5-Fluorouracil + irinotecan | 17 (6.4) | 2 (4.8) | 8 (13.8) | 6 (9.5) | 1 (1.7) | 0 |
| Capecitabine + oxaliplatin | 11 (4.2) | 0 | 0 | 1 (1.6) | 2 (3.4) | 8 (18.2) |
| Gemcitabine | 11 (4.2) | 7 (16.7) | 2 (3.4) | 0 | 2 (3.4) | 0 |
| Other single agent | 40 (15.1) | 6 (14.3) | 16 (27.6) | 6 (9.5) | 7 (12.1) | 5 (11.4) |
| Other combination | 39 (14.7) | 16 (38.1) | 14 (24.1) | 3 (4.8) | 5 (8.6) | 1 (2.3) |
| Other targeted agent | 25 (9.4) | 1 (2.4) | 6 (10.3) | 13 (20.6) | 0 | 5 (11.4) |
| 3L systemic treatment regimen, n (%) | ||||||
| None | 756 (95.5) | 159 (95.2) | 134 (89.3) | 146 (95.4) | 165 (98.8) | 152 (98.1) |
| Any | 36 (4.5) | 8 (4.8) | 16 (10.7) | 7 (4.6) | 2 (1.2) | 3 (1.9) |
1L, first-line; 2L, second-line; 3L, third-line; BTC, biliary tract cancer; ECOG, Eastern Cooperative Oncology Group; IQR, interquartile range; rwTTI, real-world time to treatment initiation.
Responses were not mutually exclusive. Only responses provided by at least 5% of the patients in the ‘Overall’ group have been included.
The median (IQR) number of 1L systemic treatment cycles was 7.6 (IQR: 5.1-10.9). The reasons for selecting 1L systemic treatment were clinical evidence supporting the treatment (52.0%-81.9%) followed by ECOG performance status of the patient (25.3%-53.3%). The median rwTTNT for 2L systemic treatment was 8.8 months (95% CI 7.1-14.9 months). BTC-related HCRU was captured (see Supplementary Figure S1, available at https://doi.org/10.1016/j.esmorw.2024.100060). Figure 2 illustrates advanced BTC treatment patterns (see Supplementary Figure S2, available at https://doi.org/10.1016/j.esmorw.2024.100060 for country-specific Sankey diagrams).
Figure 2.
Systemic treatment patterns across lines of treatment in the overall sample.
1L, first-line; 2L, second-line; 3L, third-line; BTC, biliary tract cancer. Categories (i.e. treatment lines) represented by n <3 are grouped in the ‘other single agent’ or ‘other combination’ categories depending on the number of agents received for that line of treatment.
One-third of patients (33.5%) transitioned to a 2L treatment, among whom 43.1% started 2L within 6 months of 1L discontinuation. This finding was consistent across countries (Table 2). Most commonly used regimens were 5-fluorouracil–oxaliplatin (21.5%) and capecitabine (16.6%) while 17.0% received a 2L targeted agent(s). Compared with 1L systemic treatment, greater differences between countries in the treatment regimens used for 2L systemic treatment were observed, including more use of gemcitabine monotherapy in France (16.7% of patients) compared with the four other countries (≤3.4%) and more use of 5-fluorouracil + irinotecan in Germany (13.8%), Spain (9.5%) compared with France (4.8%), Italy (1.7%), and the UK (0%).
Treatment outcomes
Seventeen patients with a partial date of death were excluded from the overall survival analysis. The median rwOS for the overall population was 13.4 months (95% CI 12.7-14.6 months; Figure 3). Results by country can be found in Supplementary Table S3, available at https://doi.org/10.1016/j.esmorw.2024.100060. The median rwOS was 14.5 months among patients treated with cisplatin–gemcitabine 1L and was numerically lower (12.2 months) with other 1L systemic treatment regimens (Table 3). Patients receiving cisplatin–gemcitabine 1L versus other 1L systemic regimens had similar age (age <65 years: 43.3% versus 41.4%) but better performance status (ECOG ≤1: 95.3% versus 65.4%) at baseline.
Figure 3.
Kaplan–Meier plot of real-world overall survival (rwOS). Sensitivity analyses were carried out to include patients with a partial date of death; results remained consistent with primary analysis (data not shown).
1L, first-line; CI, confidence interval; rwOS, real-world overall survival.
Table 3.
Real-world overall survival—subgroup analysis
| N = 775a | Events n (%) | Censored n (%) | % rwOS at 12 months (95% CI) | Median rwOS months (95% CI) |
|---|---|---|---|---|
| Primary medical specialty | ||||
| Oncologist | 356 (53.9) | 305 (46.1) | 57.1 (53.1-60.9) | 13.6 (12.8-14.7) |
| Hepatologist, gastroenterologist or internal medicine | 71 (62.3) | 43 (37.7) | 50.7 (40.6-59.9) | 12.1 (9.4-17.1) |
| 1L Treatment | ||||
| Cisplatin-gemcitabine | 189 (50.9) | 182 (49.1) | 62.2 (56.9-67.1) | 14.5 (13.416.6) |
| Non-cisplatin-gemcitibine systemic treatmentb | 238 (58.9) | 166 (41.1) | 50.5 (45.2-55.5) | 12.2 (10.9-14.3) |
| Site of primary tumorb | ||||
| Extrahepatic CCA | 144 (62.6) | 86 (37.4) | 47.6 (40.7-54.2) | 11.4 (10.1-13.0) |
| Intrahepatic CCA | 143 (52.4) | 130 (47.6) | 58.4 (52.0-64.3) | 14.5 (12.6-18.5) |
| Gallbladder cancer | 97 (53.9) | 83 (46.1) | 59.0 (51.1-66.0) | 13.8 (12.2-17.5) |
| Ampullary cancer | 36 (45.0) | 44 (55.0) | 61.9 (49.9-71.8) | 17.1 (11.1-NE) |
| ECOG performance statusc | ||||
| 0 | 77 (43.8) | 99 (56.3) | 65.0 (57.0-71.9) | 17.1 (13.8-37.8) |
| 1 | 243 (55.1) | 198 (44.9) | 57.9 (52.9-62.5) | 13.9 (12.7-16.0) |
| 2+ | 81 (68.6) | 81 (68.6) | 34.4 (25.5-43.4) | 8.7 (7.0-10.1) |
| 1L treatment initiation | ||||
| 2018-February 2020 | 68 (74.7) | 23 (25.3) | 56.0 (45.2-65.5) | 12.7 (10.4-16.5) |
| March 2020-2022 | 359 (52.5) | 325 (47.5) | 56.2 (52.2-60.0) | 13.6 (12.8-14.9) |
| TTI | ||||
| ≤1 month | 258 (61.3) | 163 (38.7) | 53.0 (48.0-57.8) | 12.8 (11.3-14.4) |
| 1 to <3 months | 143 (50.9) | 138 (49.1) | 60.6 (54.4-66.3) | 14.3 (13.0-17.3) |
| ≥3 months | 26 (35.6) | 47 (64.4) | 59.0 (42.9-72.0) | 15.1 (10.7-NE) |
| LoT | ||||
| 1L | 314 (61.0) | 201 (39.0) | 46.0 (41.3-50.4) | 11.0 (10.2-12.1) |
| 2L+d | 113 (43.5) | 147 (56.5) | 75.6 (69.6-80.5) | 17.5 (15.1-22.2) |
| 2L+e | 113 (43.5) | 147 (56.5) | 35.5 (27.0-44.1) | 8.2 (6.3-9.9) |
1L, first-line; 2L, second-line; CCA, cholangiocarcinoma; CI, confidence interval; LoT, line of therapy; NE, not estimated; rwOS, real-world overall survival; TTI, time to treatment initiation.
Seventeen patients with a partial date of death were excluded from the analysis.
Non-cisplatin–gemcitabine systemic treatment group includes those treated with other gemcitabine- or cisplatin-containing regimens, or a monotherapy.
The results for patients with ‘not available data’ in this subgroup are not shown.
Measured from 1L treatment initiation.
Measured from 2L treatment initiation.
Discussion
This study adds new data to the published real-world evidence on demographic and clinical characteristics, real-world treatment patterns, and clinical outcomes in advanced BTC patients in Europe,6,10, 11, 12 providing overall and country-level findings in a treated population.
The physicians who contributed data to this retrospective chart review were primarily medical oncologists with substantial experience treating advanced BTC and were distributed widely across different treatment settings and regions in the five countries. The resulting patient sample reflected consistent characteristics with other published European real-world studies,6,10, 11, 12 the majority of them aged ≥60 years with ECOG ≥1 at the time of advanced BTC diagnosis and most of them with advanced disease.13 Notably, there were a high proportion of patients with eCCAs, which may help explain the high frequency of cholestasis at diagnosis that was observed.
The current study explored and uncovered key reasons for a delay of 3+ months in systemic treatment initiation from diagnosis of BTC/advanced BTC. The magnitude of this treatment delay itself was consistent with results from a recent Canadian study.14 It is notable that >50% of the chart review patient sample had jaundice at the time of diagnosis, with 61.4% requiring stenting, which may have contributed to the observed delay in treatment initiation. Another interesting finding was that a substantial proportion of the delays in treatment (40.4%) were reported to be due to patient choice. This finding suggests a need for further research to understand patient choices regarding delay of treatment initiation and any potential need for informed patient and physician joint decision-making in real-world practice.
This study also offered the opportunity to assess the extent to which real-world treatment patterns are guideline driven. The chart review highlighted that close to half of the patients initiated cisplatin–gemcitabine, the ESMO recommended standard-of-care treatment as 1L therapy at the time of data capture.15,16 It is important to note that the study period was before important developments in immunotherapy, including publication of TOPAZ-1 and the 2022 approval of durvalumab in Europe as well as KEYNOTE-966 and the 2022 ESMO recommendation of pembrolizumab treatment in patients with microsatellite instability-high/mismatch repair deficient (MSI-H/dMMR) who show progression after, or intolerance to prior treatment. Therefore, this study reflects treatment patterns that likely have shifted with the introduction of these therapies, and, as such, results may serve as baseline values to which the impact of the introduction of new treatments in real-world practice can be compared in the future. That said, other European studies in the same period reported cisplatin–gemcitabine 1L use between 60%-70% in selected populations, such as in a national survey of BTC management practices in France17 or in a large European registry study of CCA patients receiving any palliative chemotherapy.6
For patients with ECOG 2+ or renal impairment, the ESMO guidelines recommend alternative 1L treatment regimens, such as gemcitabine monotherapy or a combination of gemcitabine and oxaliplatin. Indeed, our results showed that for patients with ECOG 2+, gemcitabine monotherapy increased to 20.8% (compared with 8.5% overall) while cisplatin–gemcitabine decreased to 10.8% (compared with 47.9% overall). Our chart review indicated renal impairment toxicities in 13.6% of patients but did not explore the role of renal impairment in treatment choice.
Some choices for treatments other than cisplatin–gemcitabine may also reflect consideration of comparative efficacy and safety. A recent systematic review of 33 studies (1470 patients) compared cisplatin–gemcitabine and gemcitabine plus oxaliplatin for the treatment of advanced BTC and found similar survival but higher toxicity with cisplatin–gemcitabine.18
Overall, 69.6% of patients in our chart review sample received a gemcitabine-based regimen 1L, with variation across countries in specific combinations with gemcitabine. The variety of specific gemcitabine-containing combinations were also observed in a retrospective analysis of Italian registry data and in the pan-European ENSCCA registry study in both of which the most common 1L treatment regimens were cisplatin–gemcitabine, gemcitabine–oxaliplatin, and gemcitabine monotherapy.6,11
This chart review also evaluated survival outcomes and confirmed that overall survival among this treated patient population with advanced BTC remains low (13.4 months). These survival results fall within the ranges reported in published real-world European studies that observed median OS (from 1L initiation) between 8.9 and 16.7 months, with differences in OS by age, treatment regimen, and the number of lines of treatment received.6,10,12 Based on the subgroup analyses carried out, data suggest that patients treated with cisplatin–gemcitabine 1L had numerically longer median OS relative to other regimens (14.5 versus 12.2 months, respectively). This result may have been partly due to the observed higher proportion of patients with a better performance score at 1L initiation for cisplatin–gemcitabine when compared with any other treatment regimen. This, in turn, may reflect ESMO recommendations for gemcitabine monotherapy for patients with ECOG 2 and cisplatin–gemcitabine in patients with ECOG 0 or 1.
While this study was not powered for statistical comparisons between the included countries, we note some heterogeneity of data across EU countries, including the numerically lower response and survival outcomes in the UK than in France, Germany, Italy, and Spain. This is particularly notable given that the patient population in the UK had on average a lower ECOG performance status than the patients in France, Germany, Italy, and Spain. These differences remained after excluding patients with an extended delay to the start of 1L systemic treatment. In addition, the lower OS in the UK cannot be attributed to low usage or longer time to initiation of 2L therapy in the UK, as these were similar to other countries among surviving patients. A notable difference, however, was the higher age at diagnosis in the UK and the higher proportion of patients who did not receive treatment before advanced BTC diagnosis. These data are supported by recent UK registry data in which emergency presentations of CCA leading to initial diagnosis are documented at ∼50%.19 The findings on survival, together with variations in age at diagnosis and treatment outcomes, have implications for health policy in Europe.20 Future studies should investigate if country-specific challenges in treatment, such as socioeconomic issues affecting treatment adherence and/or low awareness of treatment benefits, could be responsible for these differences, which, in turn, could inform effective country level policies.
Recent advancements in the treatment of advanced BTC patients including immunotherapy-based therapy have shown significant improvement in overall survival.21 Moreover, targeted therapy has improved survival outcomes in molecularly defined settings, such as in patients with fibroblast growth factor receptor 2 (FGFR2) fusions/rearrangements or Isocitrate dehydrogenase NADP+ 1 (IDH1) mutations. A recent multicenter European study confirmed effectiveness, activity, and safety of pemigatinib in patients affected by locally advanced or metastatic BTC with fusion or rearrangements of FGFR2 in a real-world setting.22
The primary limitation of this study was the retrospective data collection. Secondly, this study focused specifically on advanced BTC patients who received 1L systemic treatment, and, as such, any patients who did not survive the initiation of 1L treatment were excluded by design. This is difficult to estimate, but in a recent UK estimate, is as great as 50%.23 Thirdly, while efforts were made to ensure that patient charts were selected randomly by matching a randomly generated letter from the alphabet with the patient’s initials, the potential for selection bias should be noted. Fourthly, results were based on data abstracted from real-world practice and available information recorded in patient notes. As such, physicians may have used different criteria for assessing some clinical outcomes (e.g. response rates, progression), however, overall survival would not have been affected. Fifthly, data gathered from existing records may not be complete; however, a response field for recording ‘unknown/data not available’ was provided for all questions. The proportion of missing data was relatively low, typically <5%, suggesting that study results were not substantially impacted by missing data.
Conclusion
This study’s findings underscore the significant burden faced by patients with advanced BTC with a limited survival of just more than a year after 1L treatment initiation. This study highlights the need to intensify efforts in early detection and classification of BTC, the development and provision of access to new treatments, and their evaluation in real-world practice, especially in light of the recent approval of immunotherapy and targeted therapies for patients with advanced BTC.
Acknowledgements
The authors thank Iqra Syed, employee of AstraZeneca, for previous study leadership and epidemiology support for this study, along with Rachael Mann and Jovita Brantley, employees of Fortrea, for their medical writing and editorial support in the development of this manuscript. The authors are grateful for the fieldwork and data collection oversight provided by A+A Healthcare during the course of the study.
Funding
This work was supported by AstraZeneca. Employees of AstraZeneca were involved in the design and conduct of the study, management, and interpretation of the data, and review of the manuscript.
Disclosure
JB, co-author, has received advisory board fees from AstraZeneca. JMB, co-author, has received consulting fees from AstraZeneca. JS, BB, and MP, co-authors, are AstraZeneca employees, and AstraZeneca stockholders. AS and PM, co-authors and Fortrea employees, carried out the contracted study research funded by AstraZeneca.
Supplementary data
References
- 1.Forner A., Vidili G., Rengo M., Bujanda L., Ponz-Sarvisé M., Lamarca A. Clinical presentation, diagnosis and staging of cholangiocarcinoma. Liver Int. 2019;39(suppl 1):98–107. doi: 10.1111/liv.14086. [DOI] [PubMed] [Google Scholar]
- 2.Valle J.W., Kelley R.K., Nervi B., Oh D.Y., Zhu A.X. Biliary tract cancer. Lancet. 2021;397(10272):428–444. doi: 10.1016/S0140-6736(21)00153-7. [DOI] [PubMed] [Google Scholar]
- 3.Vithayathil M., Khan S.A. Current epidemiology of cholangiocarcinoma in Western countries. J Hepatol. 2022;77(6):1690–1698. doi: 10.1016/j.jhep.2022.07.022. [DOI] [PubMed] [Google Scholar]
- 4.Khan S.A., Tavolari S., Brandi G. Cholangiocarcinoma: epidemiology and risk factors. Liver Int. 2019;39(suppl 1):19–31. doi: 10.1111/liv.14095. [DOI] [PubMed] [Google Scholar]
- 5.WHO WHO mortality database. https://platform.who.int/mortality Available at. Accessed December 20, 2023.
- 6.Izquierdo-Sanchez L., Lamarca A., La Casta A., et al. Cholangiocarcinoma landscape in Europe: diagnostic, prognostic and therapeutic insights from the ENSCCA Registry. J Hepatol. 2022;76(5):1109–1121. doi: 10.1016/j.jhep.2021.12.010. [DOI] [PubMed] [Google Scholar]
- 7.Banales J.M., Marin J.J.G., Lamarca A., et al. Cholangiocarcinoma 2020: the next horizon in mechanisms and management. Nat Rev Gastroenterol Hepatol. 2020;17(9):557–588. doi: 10.1038/s41575-020-0310-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bridgewater J.A., Goodman K.A., Kalyan A., Mulcahy M.F. Biliary tract cancer: epidemiology, radiotherapy, and molecular profiling. Am Soc Clin Oncol Educ Book. 2016;35:e194–e203. doi: 10.1200/EDBK_160831. [DOI] [PubMed] [Google Scholar]
- 9.Kim Y.S., Jeong C.Y., Song H.N., et al. The efficacy of fluoropyrimidine-based adjuvant chemotherapy on biliary tract cancer after R0 resection. Chin J Cancer. 2017;36(1):9. doi: 10.1186/s40880-017-0182-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lagenfelt H., Blomstrand H., Elander N.O. Real-world evidence on palliative gemcitabine and oxaliplatin (GemOx) combination chemotherapy in advanced biliary tract cancer. Cancers (Basel) 2021;13(14):3507. doi: 10.3390/cancers13143507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rizzo A., Salati M., Frega G., et al. Second-line chemotherapy in elderly patients with advanced biliary tract cancer: a multicenter real-world study. Medicina (Kaunas) 2022;58(11):1543. doi: 10.3390/medicina58111543. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Thol F., Gairing S.J., Czauderna C., et al. Outcomes in patients receiving palliative chemotherapy for advanced biliary tract cancer. JHEP Rep. 2022;4(3) doi: 10.1016/j.jhepr.2021.100417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Neuzillet C., Artru P., Assenat E., et al. Optimizing patient pathways in advanced biliary tract cancers: recent advances and a French perspective. Target Oncol. 2023;18(1):51–76. doi: 10.1007/s11523-022-00942-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Seung S.J., Saherawala H., Syed I., Shephard C., Clouthier D.L., Chen E. Real-world treatment patterns and survival outcomes for treated biliary tract cancer patients using administrative databases in Ontario. J Gastrointest Oncol. 2023;14(4):1806–1816. doi: 10.21037/jgo-23-155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.National Comprehensive Cancer Network. NCCN Guidelines Version 3.2021 . 2013. Biliary tract cancers: extrahepatic cholangiocarcinoma. Available at http://nccn.ir/Guidelines/Summary/hepatobiliary.html#pf29. Accessed December 20, 2023. [Google Scholar]
- 16.Vogel A., Bridgewater J., Edeline J., et al. Biliary tract cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol. 2023;34(2):127–140. doi: 10.1016/j.annonc.2022.10.506. [DOI] [PubMed] [Google Scholar]
- 17.Delaye M., Assenat E., Boleslawski E., et al. État des lieux des pratiques de prise en charge des cancers des voies biliaires en France: résultats de l’enquête nationale ACABi. Bull Cancer. 2022;109(11s):11S3–11S10. doi: 10.1016/S0007-4551(22)00463-5. [DOI] [PubMed] [Google Scholar]
- 18.Fiteni F., Nguyen T., Vernerey D., et al. Cisplatin/gemcitabine or oxaliplatin/gemcitabine in the treatment of advanced biliary tract cancer: a systematic review. Cancer Med. 2014;3(6):1502–1511. doi: 10.1002/cam4.299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zalin-Miller A., Jose S., Knott C., et al. Regional variation in routes to diagnosis of cholangiocarcinoma in England from 2006 to 2017. World J Gastroenterol. 2023;29(24):3825–3842. doi: 10.3748/wjg.v29.i24.3825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Berchet C., Dedet G., Klazinga N., Colombo F. Inequalities in cancer prevention and care across Europe. Lancet Oncol. 2023;24(1):10–11. doi: 10.1016/S1470-2045(22)00746-X. [DOI] [PubMed] [Google Scholar]
- 21.Oh D.Y., He A.R., Qin S., et al. Updated overall survival (OS) from the phase III TOPAZ-1 study of durvalumab (D) or placebo (PBO) plus gemcitabine and cisplatin (+ GC) in patients (pts) with advanced biliary tract cancer (BTC) Ann Oncol. 2022;33(suppl 7):S565–S566. [Google Scholar]
- 22.Parisi A., Delaunay B., Pinterpe G., et al. Pemigatinib for patients with previously treated, locally advanced or metastatic cholangiocarcinoma harboring FGFR2 fusions or rearrangements: a joint analysis of the French PEMI-BIL and Italian PEMI-REAL cohort studies. Eur J Cancer. 2024;200 doi: 10.1016/j.ejca.2024.113587. [DOI] [PubMed] [Google Scholar]
- 23.Accessed January 15, 2024. https://ammf.org.uk/ Accessed January 15, 2024. AMMF. Available at.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.