Clinical outcomes for previously treated patients with advanced biliary tract cancer: a meta-analysis
Abstract
Aim: A systematic review and meta-analysis were performed to evaluate the efficacy of treatments for previously treated advanced biliary tract cancer (BTC) patients. Materials & methods: Databases were searched for studies evaluating treatments for advanced (unresectable and/or metastatic) BTC patients who progressed on prior therapy. Pooled estimates of objective response rate (ORR), median overall survival (OS) and median progression-free survival (PFS) were calculated using random effects meta-analysis. Results: Across 31 studies evaluating chemotherapy or targeted treatment regimens in an unselected advanced BTC patient population, pooled ORR was 6.9%, median OS was 6.6 months and median PFS was 3.2 months. Conclusion: The efficacy of conventional treatments for previously treated advanced BTC patients is poor and could be improved by novel therapies.
Plain language summary – Clinical value of treatments for previously treated patients with advanced biliary tract cancer
What is this article about? Most patients with biliary tract cancer are identified with advanced disease, and almost all go through a worsening of the disease after their first treatment. For patients who go on to receive their next treatment, current guidelines are unclear regarding the best treatment choice. Therefore, we examined the available medical literature and performed an analysis of multiple studies to calculate overall estimates of the clinical value of standard treatments for these patients. Our goal was to develop a benchmark against which to compare the clinical value of new treatments that are currently being assessed in clinical trials. What were the results? We identified 31 studies assessing standard treatments (involving chemotherapy or molecularly targeted treatments) in previously treated advanced biliary tract cancer patients. Across these studies, the objective tumor response rate was 6.9%, median overall survival was 6.6 months and median progression-free survival was 3.2 months. What do the results of the study mean? These results indicate that there is limited clinical value of standard treatments for patients with advanced biliary tract cancer whose disease worsened after first treatment. This medical need could potentially be met by new treatments, such as immunotherapies that restore the immune system's ability to attack cancer cells and thereby prolong patient survival.
Tweetable abstract
A synthesis of evidence from clinical studies indicates the need for more effective therapies for advanced biliary tract cancer patients whose disease progressed on prior therapy. #cholangiocarcinoma #bileductcancer
Biliary tract cancer (BTC), which accounts for approximately 10–25% of all hepatobiliary malignancies, comprises a group of often lethal malignancies that originate in the bile ducts, gallbladder or ampulla of Vater [1]. The incidence of BTC differs by geographic region and is related to variations in the distribution of risk factors including parasitic infections, primary sclerosing cholangitis, biliary duct cysts, hepatolithiasis and toxin exposure [1,2]. Although the incidence of BTC is low in Western countries, it is high in Southeast Asian countries such as Thailand, where the rate is as high as 113 per 100,000 for men and 50 per 100,000 for women [2,3]. Furthermore, while the incidence of gallbladder cancer is declining globally, the incidence of bile duct cancer (i.e., cholangiocarcinoma) is rising [4–7]. Owing to the poor prognosis of BTC, its incidence and mortality rates are comparable [1]. Varying by disease stage, 5-year survival rates for intrahepatic cholangiocarcinoma, extrahepatic cholangiocarcinoma and gallbladder cancer range from 2 to 15%, 2 to 30% and 2 to 70%, respectively [8].
As BTC often produces nonspecific symptoms, most patients present at an advanced stage of disease, with more than 80% of cases considered unresectable at the time of diagnosis [4,8]. The standard first-line systemic chemotherapy regimen for advanced BTC is the combination of cisplatin and gemcitabine [9], which was found to significantly improve overall survival (OS) and progression-free survival (PFS) compared with gemcitabine alone [10]. However, nearly all BTC patients exhibit disease progression after first-line therapy, and there is presently no strong evidence that the 30–35% of patients who go on to receive second-line therapy experience meaningful improvements in OS [11]. Indeed, an earlier systematic review of studies evaluating second-line treatments for patients with advanced BTC published in 2014 concluded that the quality of the evidence base was too poor to recommend a second-line chemotherapy schedule for this patient population [12]. As such, current BTC treatment guidelines note uncertainty regarding the best treatment options after first-line therapy and encourage patient participation in clinical trials [9,13–17].
Promisingly, the therapeutic options for BTC patients are rapidly expanding due to advances in our understanding of the molecular landscape of BTC and the development of new therapies targeting specific molecular aberrations [11]. Compared with other cancer types, BTC is associated with a high prevalence of oncogenic drivers, with around 40–50% of BTC patients exhibiting tumor molecular alterations, including mutations in IDH1, FGFR2, ERBB2/HER-2, BRAF or NTRK, that are targetable by specific inhibitors [18]. For the other 50–60% of BTC patients without targetable molecular alterations, however, immunotherapies have the potential to improve tumor response to therapy and prolong patient survival. In particular, several clinical trials evaluating the efficacy of immune checkpoint inhibitors delivered in combination with chemotherapy or targeted therapy among previously treated BTC patients are currently ongoing [18].
A recent systematic review published in 2022 concluded that a lack of overlap among investigated treatments combined with between-study heterogeneity in study and patient characteristics precluded investigation of the comparative efficacy of different second-line treatment regimens for advanced BTC patients via network meta-analysis [19]. Although determination of the relative efficacy of available second-line treatment options may not be possible at present, pooled estimates of the overall efficacy of second-line chemotherapy or targeted therapy regimens for advanced BTC patients could help establish a benchmark against which to compare the efficacy of emerging immunotherapies in recently completed and ongoing trials. Therefore, the objectives of this study were to perform a systematic review and meta-analysis to 1) provide an overview of interventional studies evaluating the efficacy, safety and patient-reported outcomes of treatments for advanced (i.e., unresectable or metastatic) BTC patients who progressed on prior therapy and 2) calculate pooled estimates of the efficacy of treatment regimens involving chemotherapy or targeted therapies within an unselected advanced BTC patient population.
Methods
A systematic review was performed and reported in accordance with Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) 2020 guidelines [20] and guided by predefined population, intervention, comparator, outcome, study design, time (PICOST) criteria (Table 1). The review was not registered in a systematic review registry. Because standard-of-care therapy for patients with advanced BTC who were previously treated for advanced disease is not well defined, interventions of interest were any pharmacologic treatment licensed by the US FDA or European Medicines Agency for any indication, including off-label treatments. To reflect current patient populations and clinical practice, publications were excluded if they were published before 2000.
| Criteria | Inclusion criteria | Exclusion criteria |
|---|---|---|
| Population | Adults (≥18 years) with advanced (unresectable and/or metastatic) biliary adenocarcinoma, including intrahepatic or extrahepatic cholangiocarcinoma and gallbladder cancer • Previously treated for advanced disease • Performance status of 0–1 (or equivalent) • Recurrent disease when stage not specified | Populations consisting exclusively of patients with one of the following characteristics: • Ampulla of Vater cancer • Previously treated with anti-PD1/PD-L1 agents • Performance status of ≥2 (or equivalent) • Stage I or II disease • Central nervous system metastasis |
| Interventions/comparators | Any pharmacologic treatment licensed by the US FDA or European Medicines Agency for any indication, including off-label treatments | NA |
| Outcomes | Efficacy outcomes: • Objective response rate (and numbers of patients with complete response or partial response when available) • Median overall survival • Median progression-free survival Safety outcomes: • Any-cause and treatment-related adverse events • Any-cause and treatment-related grade 3–5 adverse events • Any-cause and treatment-related serious adverse events • Discontinuation due to adverse events • Deaths due to adverse events Patient-reported outcomes (e.g., EQ-5D, EORTC QLQ-C30) | NA |
| Study design | • Randomized controlled trials • Nonrandomized trials • Single-arm trials | • Animal or in vitro studies • Case reports/series • Editorials, commentaries, letters, reviews • Pharmacokinetics or pharmacodynamics studies |
| Time | From 2000 onward | NA |
| Language | English | NA |
Search strategy
Database searches were conducted in Embase, MEDLINE and Cochrane Central Register of Controlled Trials (CENTRAL) via OVID. The search strategies included keywords and subject headings as appropriate for each database describing the population of interest (i.e., patients with advanced BTC). Embase and MEDLINE search strategies additionally employed the Scottish Intercollegiate Guidelines Network's search filters for randomized controlled trials (RCTs; www.sign.ac.uk/methodology/filters.html), which were modified to also retrieve single-arm trials [21]. To identify trials not yet been published but potentially eligible for inclusion, searches of American Society of Clinical Oncology (2019–2021) and European Society for Medical Oncology (2019–2020) conference proceedings were performed using the Northern Lights database. In addition, the US National Institute of Health Clinical Trials Registry (ClinicalTrials.gov) was searched to identify trials with available results that had not yet been published. Embase, MEDLINE, CENTRAL and Northern Lights searches were conducted on 29 June 2021, and the ClinicalTrials.gov search was conducted on 9 August 2021. Complete search strategies for each database are provided in Supplementary Tables 1–6.
Study selection & quality assessment
Title/abstract and full-text screening were performed based on predefined PICOST criteria (Table 1). After studies were selected for inclusion, the Cochrane Collaboration's Risk of Bias tool version 2 was used to assess the risk of bias of RCTs [22], and the Newcastle-Ottawa Scale for Cohort Studies was used to assess the quality of single-arm and nonrandomized trials [23]. For the Newcastle-Ottawa scale, possible scores ranged from 0 to 9, with higher scores reflective of higher quality. Title/abstract, full-text screening and study quality assessment were performed by two independent reviewers. Discrepancies between reviewers were resolved by discussion and the involvement of a third reviewer, if needed. To assess potential publication bias, visual inspection of a funnel plot and Egger's test among studies reporting objective response rate (ORR) were performed.
Data extraction
Citations reporting on overlapping samples of patients were carefully linked to each other to avoid double-counting patients. Available data on study, patient and treatment characteristics and reported outcomes were extracted from the included studies by two independent reviewers. Discrepancies between reviewers were resolved by discussion and the involvement of a third reviewer, if needed. Outcomes of interest included ORR (typically defined as the proportion of patients with a complete or partial response to treatment), OS (typically defined as the time from treatment to death), PFS (typically defined as the time from treatment to disease progression or death), safety outcomes and patient-reported outcomes. Where possible, ORR and the number of responders was taken directly from the reported study results. For some studies [24–26], we recalculated ORR using the intention-to-treat sample size as the denominator instead of the number of evaluable patients, as the published evaluable sample excluded patients who progressed prior to the first scan. If ORR was not directly available, the number of total responders was determined by summing the numbers of complete and partial responders. Conversely, if the number of responders was not reported, this value was estimated from the reported ORR and number of patients in the trial arm. For trials in which partial response but not complete response was reported, it was assumed that the number of complete responders was zero. Kaplan–Meier curves for OS and PFS provided in the study publications were digitized using DigitizeIt (www.digitizeit.xyz/), and pseudo-individual patient level data were estimated using an algorithm by Guyot et al. [27].
Meta-analysis
Whereas the systematic review was designed to identify studies evaluating all second-line treatments for patients with advanced BTC, studies were included in the meta-analysis only if they evaluated chemotherapy or targeted treatment regimens (not involving immunotherapy agents) delivered in a biomarker-unselected advanced BTC patient population. Due to inherent heterogeneity among included studies considering treatments, dose intensities, study design, length of follow-up, eligible populations and outcome measurements, a random effects model was the primary approach used in meta-analyses to synthesize the overall estimates of efficacy.
For ORR, we employed the Freeman–Tukey double-arcsine transformation to normalize and stabilize the variance in raw response rates [28] from each trial arm and then back-transformed the synthesized results such that the final estimates were reported as proportions [29]. Inverse-variance weighted random effects meta-analysis was carried out using the DerSimonian-Laird moment method to estimate between-study variance (τ2) [30]. CIs were calculated using the Clopper–Pearson method. The analysis was performed using SAS version 9.4. Reported results include sample size, pooled estimate of ORR with 95% CI, I2 and τ2 statistics calculated using previously reported formulas [31], p-value from Cochran's Q test for heterogeneity and a forest plot.
For OS and PFS, pseudo-individual patient-level data obtained from digitized Kaplan–Meier curves from each trial arm were pooled using methods described by Combescure et al. [32]. Conditional survival probabilities at each time point were arcsine transformed and pooled, and then used to evaluate summary survival probabilities. Median survival times were derived from summary survival curves assuming a linear interpolation of survival between time points. Pooled survival probabilities were calculated at every half month of follow-up time until no included trial had participants at risk. CIs were calculated using Greenwood's formula. Pooling of survival curves was performed using the metasurv package in R version 4.0.1. Reported results include sample size; median survival time with 95% CI; survival rates at 6, 12 and 24 months; I2 statistics calculated using a previously reported formula [32]; p-values from Cochran's Q tests for heterogeneity and pooled survival curves.
Results
Systematic review
Study selection
Out of a total of 4578 citations identified through searches of Embase, MEDLINE, CENTRAL, Northern Lights and ClinicalTrials.gov, 53 citations representing 39 unique studies were selected for inclusion in the systematic review (Figure 1).
Flow diagram depicting the study search and selection processes.
Summary of study characteristics
The 39 studies consisted of seven RCTs, 31 single-arm trials and one nonrandomized trial (Table 2). A total of 38 unique treatment regimens were evaluated, the most common being regorafenib (n = 3 studies), gemcitabine + cisplatin (n = 2 studies), S-1 (n = 2 studies), irinotecan + capecitabine (XELIRI/CAPIRI; n = 2 studies) and pembrolizumab (n = 2 studies). Most studies were conducted in Japan (n = 8 studies), the USA (n = 7 studies), multiple countries (n = 7 studies) or South Korea (n = 5 studies). In addition, most studies were small, with a median of 30 patients per trial arm (range: 6–104). Detailed study, treatment and patient characteristics are provided in Supplementary Tables 7–9.
| Trial | Study design | Country | Trial arm | Treatment | n | Ref. |
|---|---|---|---|---|---|---|
| Bengala et al. (2010) | Single-arm | NR | 1 | Sorafenib | 26 | [33] |
| Sasaki et al. (2011) | Single-arm | Japan | 1 | Gemcitabine + cisplatin | 20 | [34] |
| Rohrberg et al. (2012) | Single-arm | Denmark | 1 | Erlotinib + bevacizumab | 16 | [35] |
| Sasaki et al. (2012) | Single-arm | Japan | 1 | S-1 | 22 | [36] |
| Yi et al. (2012) | Single-arm | Multinational | 1 | Sunitinib | 56 | [37] |
| Sinn et al. (2013) | Single-arm | Germany | 1 | Oxaliplatin + natrium folinate + 5-fluorouracil | 6 | [38] |
| Suzuki et al. (2013) | Single-arm | Japan | 1 | S-1 | 40 | [39] |
| Hwang et al. (2015) | Single-arm | South Korea | 1 | mFOLFOX3 (oxaliplatin + 5-fluorouracil + leucovorin) | 30 | [40] |
| Cereda et al. (2016) | RCT | Italy | 1 | Capecitabine | 26 | [41] |
| 2 | Capecitabine + mitomycin | 29 | ||||
| Goyal et al. (2017) | Single-arm | USA | 1 | Cabozantinib | 19 | [42] |
| Kobayashi et al. (2017) | Single-arm | Japan | 1 | Gemcitabine + S-1 | 41 | [43] |
| Shroff et al. (2017) | Single-arm | USA | 1 | Pazopanib + trametinib | 25 | [44] |
| Arkenau et al. (2018) | Single-arm | Multinational | 1 | Pembrolizumab + ramucirumab | 26 | [45] |
| Ikeda et al. (2018) | Single-arm | Japan | 1 | Trametinib | 20 | [46] |
| Javle et al. (2018) | Single-arm | Multinational | 1 | Infigratinib | 61 | [47] |
| Larsen et al. (2018) | Single arm | Denmark | 1 | Gemcitabine + irinotecan + bevacizumab + capecitabine | 48 | [48] |
| Matsuyama et al. (2018) | Single-arm | Japan | 1 | Gemcitabine + cisplatin | 27 | [49] |
| Zheng et al. (2018) | RCT | China | 1 | XELIRI (irinotecan + capecitabine) | 30 | [50] |
| 2 | Irinotecan | 30 | ||||
| Kim et al. (2019) | Single-arm | South Korea | 1 | Binimetinib + capecitabine | 34 | [51] |
| Kim et al. (2019) | Single-arm | South Korea | 1 | XELOX (capecitabine + oxaliplatin) | 50 | [52] |
| Park et al. (2019) | Single-arm | Multinational | 1 | Erdafitinib | 17 | [53] |
| Sun et al. (2019) | Single-arm | USA | 1 | Regorafenib | 43 | [54] |
| Belkouz et al. (2020) | Single-arm | Netherlands | 1 | FOLFIRINOX (fluorouracil + leucovorin + irinotecan + oxaliplatin) | 30 | [55] |
| Chakrabarti et al. (2020) | Single-arm | USA | 1 | Trifluridine/tipiracil | 27 | [56] |
| Demols et al. (2020) | RCT | Belgium | 1 | Regorafenib | 33 | [57] |
| 2 | Placebo | 33 | ||||
| Kim et al. (2020) | Single-arm | USA | 1 | Nivolumab | 54 | [26] |
| Kim et al. (2020) | Single-arm | USA | 1 | Regorafenib | 39 | [25] |
| Klein et al. (2020) | Single-arm | Australia | 1 | Nivolumab + ipilimumab | 33 | [58] |
| Piha-Paul et al. (2020) | Single-arm | Multinational | 1 | Pembrolizumab | 24 | [59] |
| Piha-Paul et al. (2020) | Single-arm | Multinational | 1 | Pembrolizumab | 104 | [59] |
| Ueno et al. (2020) | Single-arm | Japan | 1 | Lenvatinib | 26 | [60] |
| Choi et al. (2021) | RCT | South Korea | 1 | mFOLFOX (oxaliplatin + 5-fluorouracil) | 59 | [61] |
| 2 | mFOLFIRI (irinotecan + 5-fluorouracil) | 59 | ||||
| Lamarca et al. (2021) | RCT | UK | 1 | Active symptom control | 81 | [62] |
| 2 | Active symptom control + mFOLFOX | 81 | ||||
| Ma et al. (2021) | Single-arm | USA | 1 | Paclitaxel injection concentrate for nano-dispersion | 20 | [63] |
| Okano et al. (2021) | Single-arm | Japan | 1 | Axitinib | 19 | [64] |
| Ramaswamy et al. (2021) | RCT | India | 1 | CAPIRI (capecitabine + irinotecan) | 49 | [65] |
| 2 | Irinotecan | 49 | ||||
| Villaneuva et al. (2021) | Nonrandomized trial | Multinational | 1 | Pembrolizumab + lenvatinib | 31 | [66] |
| Woodford et al. (2021) | Single arm | Australia | 1 | Capecitabine + nab-paclitaxel | 10 | [67] |
| Yoo et al. (2021) | RCT | South Korea | 1 | Liposomal irinotecan + flurouracil + leucovorin | 88 | [68] |
| 2 | Flurouracil + leucovorin | 86 |
Summary of reported outcomes
Efficacy outcomes
Of the 36 trials reporting ORR or for which the ORR was calculable, ORR ranged from 0 [24,34,41,42,56] to 47% [57] (Supplementary Table 10). Of the 23 trials reporting OS, median OS ranged from 4.4 [53] to 13.5 months [40]. Of the 21 trials reporting PFS, median PFS ranged from 1.6 [36,53] to 6.2 months [45] (Supplementary Table 11).
Safety outcomes
Of the 14 trials reporting any-cause overall, grade 3–4 or serious adverse events, rates ranged from 82 [55] to 100% [46,48,56,65], from 30 [55] to 77.3% [54] and from 29 [57] to 69.2% [25], respectively (Supplementary Table 12). Of the 14 trials reporting treatment-related overall, grade 3–4, or serious adverse events, rates ranged from 54.8 [68] to 100% [57,59], from 11.1 [56] to 89% [42] and from 16.7 [45] to 37.7% [60], respectively. Of the 26 trials reporting discontinuation or death due to adverse events, rates ranged from 0 [47,55] to 24% [41] and from 0 [25,26,34,36,40,42,44,47,48,50,52,55,61,62,64,68] to 6.7% [45], respectively.
Patient-reported outcomes
Three trials reported patient-reported outcomes (Supplementary Table 13). One single-arm trial reported patients' baseline health-related quality of life [66], and two RCTs reported no significant difference in health-related quality of life between study arms after treatment [54,55].
Study quality assessment
Five RCTs were considered to have some concerns of bias [24,41,54,55,61], and two RCTs were considered to have a low risk of bias [44,50] (Supplementary Figure 1). The single-arm trials had quality scores of 4 [48], 5 [33,34,38,43,45,46,49,57,67] or 6 [25,26,35–37,39,40,42,47,51–53,56,58–60,62,65,66,68], with lower scores due to being single-center studies, lacking information on duration of follow-up, and/or having a high rate of loss to follow-up. The nonrandomized trial had a quality score of 9 [64] (Supplementary Table 14).
To assess potential publication bias in the literature, a funnel plot was generated and Egger's test was performed. Neither visual inspection of the funnel plot (Figure 2) nor Egger's test showed evidence of publication bias (intercept: 0.43; 95% CI: -1.57 to 2.44; p = 0.67).
A funnel plot of the 34 trial arms included in the meta-analysis of objective response rate was symmetrical. An accompanying Egger's test showed no evidence of publication bias (intercept: 0.43; 95% CI: -1.57 to 2.44; p = 0.67).
Meta-analysis
To evaluate the efficacy of treatment regimens involving chemotherapy or targeted therapy within an unselected population of patients with advanced BTC who progressed on prior therapy, 31 of the 39 trials identified in the systematic review were included in random-effects meta-analysis of ORR, OS or PFS. Two studies were excluded from the meta-analysis because they exclusively enrolled patients with FGFR alterations [57,60], and six studies were excluded because they evaluated treatment regimens involving immunotherapy agents [26,36,39,64,68]. In addition, two trial arms of nonactive therapies (i.e., placebo [24] and active symptom control [61]) were excluded.
The random effects pooled estimate of ORR from 34 trial arms (29 unique trials) was 6.9% (95% CI: 4.8–9.2%; Figure 3), with evidence of a moderate degree of heterogeneity in intervention effects among studies [31]. The pooled median OS from 26 trial arms (22 unique trials) was 6.6 months (95% CI: 5.9–7.1; Figure 4), and 6-, 12- and 24-month OS rates were 54.1 (95% CI: 49.6–59.0), 24.8 (95% CI: 20.7–29.6) and 2.4% (95% CI: 1.3–4.4), respectively (Table 3 & Supplementary Table 15), with evidence of a low degree of heterogeneity in intervention effects among studies. The pooled median PFS from 24 trial arms (20 unique trials) was 3.2 months (95% CI: 2.8–3.8; Figure 5), and 6-, 12- and 24-month PFS rates were 25.8 (95% CI: 20.3–32.7), 7.0 (95% CI: 4.5–10.8) and 0.1% (95% CI: 0.0–0.6), respectively (Table 3 & Supplementary Table 16), with evidence of a low degree of heterogeneity in intervention effects among studies.
‘Responders’ denotes the number of patients with an objective response (complete response + partial response). ‘Total’ denotes the number of patients who received at least one dose of treatment.
OR: Objective response.
Grey lines represent Kaplan–Meier estimates of overall survival for each trial arm. Black squares represent the end of follow-up for each trial. The thick black line represents the random effects pooled overall survival estimate with 95% CI (dashed lines). The p-value was derived from Cochran's Q test of heterogeneity (I2 test statistic).
| OS | PFS | |
|---|---|---|
| Number of studies | 22 | 20 |
| Number of arms | 26 | 24 |
| Number (%) of OS events | 899 (84.3) | 892 (90.5) |
| Person-months† | 8404.9 | 4187.8 |
| Event rate per 100 person-months | 10.7 | 21.3 |
| Median survival (months)‡ | 6.6 | 3.2 |
| 95% CI for median OS‡ | (5.9–7.1) | (2.8–3.8) |
| OS rate at 6 months, %‡ (95% CI) | 54.1 (49.6–59.0) | 25.8 (20.3–32.7) |
| OS rate at 12 months, %‡ (95% CI) | 24.8 (20.7–29.6) | 7.0 (4.5–10.8) |
| OS rate at 18 months, %‡ (95% CI) | 8.4 (5.7–12.4) | 1.3 (0.6–2.9) |
| OS rate at 24 months, %‡ (95% CI) | 2.4 (1.3–4.4) | 0.1 (0.0–0.6) |
| OS rate at 30 months, %‡ (95% CI) | 0.4 (0.1–1.2) | 0.0 (0.0–0.3) |
| OS rate at 36 months, %‡ (95% CI) | 0.1 (0.0–0.4) | – |
| OS rate at 42 months, %‡ (95% CI) | 0.0 (0.0–0.1) | – |
| OS rate at 48 months, %‡ (95% CI) | 0.0 (0.0–0.0) | – |
Grey lines represent Kaplan–Meier estimates of progression-free survival for each trial arm. Black squares represent the end of follow-up for each trial. The thick black line represents the random effects pooled progression-free survival estimate with 95% CI (dashed lines). The p-value was derived from Cochran's Q test of heterogeneity (I2 test statistic).
Discussion
The objectives of this systematic review and meta-analysis were to 1) identify evidence on the efficacy, safety and patient-reported outcomes of treatments for patients with advanced BTC who experienced disease progression on prior therapy and 2) estimate the pooled efficacy of treatment regimens involving chemotherapy or targeted therapy in an unselected advanced BTC patient population. We identified a total of 39 randomized controlled, nonrandomized and single-arm trials investigating 38 unique treatment regimens. However, consistent with the findings of a previous systematic review [19], a lack of overlap among investigated treatment regimens precluded evaluation of the relative efficacy of particular interventions. Therefore, to obtain pooled estimates of the efficacy of treatment regimens involving chemotherapy or targeted therapy across trials, random effects meta-analyses were performed. The pooled estimate of ORR was 6.9% (95% CI: 4.8–9.2%), pooled median OS was 6.6 months (95% CI: 5.9–7.1) and pooled median PFS was 3.2 months (95% CI: 2.8–3.8), reflecting the poor outcomes of BTC patients who go on to receive second-line therapy or beyond.
These results suggest that the efficacy of treatment regimens involving chemotherapy or targeted therapy for patients with advanced BTC who progressed on prior therapy has not improved over the last decade, as a 2014 meta-analysis of phase II clinical trials of second-line chemotherapy regimens reported a weighted mean ORR of 8.2% (95% CI: 3.9–12.4), median OS of 6.6 months (95% CI: 5.1–8.1) and median PFS of 2.8 months (95% CI: 2.1–3.5) [12]. The pooled estimates of ORR, OS and PFS calculated in the present study are similar to those reported in the previous meta-analysis, with substantial overlap in CIs between analyses, suggesting a lack of meaningful change over time.
Given the persistently poor outcomes of patients with advanced BTC, more research is needed to identify and evaluate new treatments that can improve patient prognosis while maintaining acceptable safety and quality of life. In particular, immunotherapy agents targeting specific immune checkpoint proteins, such as PD-1/PD-L1 or CTLA-4, are promising routes [11,63], although further research is necessary to identify biomarkers (e.g., tumor microsatellite status, mutational burden, PD-L1 expression) that can predict which patients will respond best to these agents [11,18,63]. In addition, given the rarity of BTC and the low proportion of patients with advanced BTC who go on to receive second-line therapy or beyond, clinical trials for this population could benefit from greater opportunities for patient enrollment afforded by multi-institutional studies and international collaboration [69].
The results of this study may be of value to several stakeholders including health technology assessment agencies, policymakers, oncologists and cancer patients, oncology drug developers and medical researchers. Health technology assessment agencies and policymakers could use the results as clinical benchmarks to contextualize the efficacy of novel treatments in second-line or beyond settings. Oncologists could use this information in the medical decision-making process and to communicate with their patients. Oncology drug developers could use these results to define statistical hypotheses and perform power calculations for future clinical trials. In addition, the use of a state-of-the-art method to perform meta-analysis using published and digitized survival curves allows the application of this technique to future studies of BTC or other cancers. In particular, medical researchers could use these results for indirect treatment comparisons, facilitating the comparative evaluation of novel therapies against the existing standard of care.
Some limitations of this systematic review and meta-analysis should be acknowledged. As interventions of interest were any licensed pharmacologic treatment for any indication rather than guideline-recommended treatments for patients with advanced BTC who progressed on prior therapy, the pooled estimates of efficacy may not precisely reflect those achieved in current clinical practice for this patient population. In addition, as with any systematic review or meta-analysis, our results are limited by the availability of published data, as some studies fail to be published, although a funnel plot and Egger's test showed no evidence of publication bias. Moreover, three out of the 39 studies included in this systematic review were published as conference abstracts, which report limited information and may not be fully peer-reviewed; therefore, their results should be interpreted with caution. To overcome some of these limitations, this systematic review involved highly sensitive searches of the peer-reviewed literature, recent conferences and a clinical trial registry to identify unpublished trials with results available. In addition, the review process was guided by predefined eligibility criteria, and data quality was ensured through the involvement of two independent reviewers in the study selection and data extraction processes.
Conclusion
Overall, the findings of this systematic review and meta-analysis indicate that the efficacy of treatment regimens involving chemotherapy or targeted therapy for advanced BTC patients who progressed on prior therapy remains poor and that new treatment approaches, such as immunotherapies, are needed to improve patient prognosis. In anticipation of the results of ongoing clinical trials evaluating the efficacy of immunotherapy agents in previously treated advanced BTC patients, the pooled estimates of the overall efficacy of current chemotherapy or targeted therapy regimens calculated in this study could help establish a benchmark against which to compare the efficacy of emerging immunotherapies for this patient population.
Most patients with biliary tract cancer (BTC) present at an advanced disease stage, and nearly all progress after first-line therapy. For patients who go on to receive second-line therapy or beyond, current guidelines note uncertainty regarding the best treatment options.
A systematic review and meta-analysis were performed to provide a clinical overview of the efficacy of treatments for advanced BTC patients who progressed on prior therapy and to calculate pooled estimates of the efficacy of treatment regimens involving chemotherapy or targeted therapies within an unselected advanced BTC patient population.
Embase, MEDLINE, Cochrane Central Register of Controlled Trials, ClinicalTrials.gov and conference abstracts were searched for interventional studies evaluating treatments for patients with advanced (unresectable and/or metastatic) BTC, including intrahepatic or extrahepatic cholangiocarcinoma and gallbladder cancer, who progressed on prior therapy.
Pooled estimates of objective response rate, median overall survival and median progression-free survival across studies were calculated using random effects meta-analysis.
A total of 39 studies were included in the systematic literature review, of which 31 evaluated chemotherapy or targeted treatment regimens in an unselected advanced BTC patient population and reported sufficient outcome data to be included in the meta-analysis.
Across studies, pooled objective response rate was 6.9% (95% CI: 4.8–9.2%), median overall was 6.6 months (95% CI: 5.9–7.1) and median progression-free survival was 3.2 months (95% CI: 2.8–3.8).
Overall, the results indicate that the efficacy of conventional treatments for patients with advanced biliary tract cancer who progressed on prior therapy remains poor. This unmet need could potentially be addressed by novel treatments, such as immunotherapies.
Supplementary data
To view the supplementary data that accompany this paper please visit the journal website at: www.futuremedicine.com/doi/suppl/10.2217/fon-2023-0006
Author contributions
Conception and design (all authors); acquisition and collection of data (GE Fox, AM Frederickson); statistical analyses (LA Abderhalden, T Grira, W Malbecq); interpretation of data (all authors); draft manuscript (GE Fox, KG Akers, AM Frederickson); critical review of manuscript (all authors). All aspects of the systematic review (e.g., searches, screening, data extraction) were conducted by PRECISIONheor. Analytical decisions were made by study investigators. All authors have read and approved the final version of this manuscript.
Acknowledgments
The authors thank Adekemi Adeyemi (PRECISIONheor), Kevin Bouliane (PRECISIONheor) and Kevin Coady (PRECISIONheor) for their contributions to the systematic review.
Financial disclosure
MM Amonkar, LA Abderhalden, A Gozman, U Malhotra and W Malbecq are employees of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc. (NJ, USA) and own stock in Merck & Co., Inc. GE Fox, AM Frederickson and KG Akers are employees of PRECISIONheor, a healthcare research consultancy that received funding from Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Competing interests disclosure
The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Writing disclosure
No writing assistance was utilized in the production of this manuscript.
Open access
This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/
Papers of special note have been highlighted as: • of interest; •• of considerable interest
References
- 1. . Risk factors for cholangiocarcinoma. Hepatology 54(1), 173–184 (2011).
- 2. . Biliary tract cancers: epidemiology, molecular pathogenesis and genetic risk associations. Chin. Clin. Oncol. 5(5), 4 (2016).
- 3. . Incidence and mortality of cancers of the biliary tract, gallbladder, and liver by sex, age, race/ethnicity, and stage at diagnosis: United States, 2013 to 2017. Cancer Epidemiol. Biomark. Prev. 30(9), 1607–1614 (2021).
- 4. . Biliary tract cancer. Lancet 397(10272), 428–444 (2021). •• Serves as a comprehensive and authoritative review of biliary tract cancer (BTC), epidemiology and risk factors, anatomical and histopathological classification, symptoms and diagnosis, surgery and treatment and management of advanced disease.
- 5. Global trends in intrahepatic and extrahepatic cholangiocarcinoma incidence from 1993 to 2012. Cancer 126(11), 2666–2678 (2020).
- 6. Epidemiology of cholangiocarcinoma; United States incidence and mortality trends. Clin. Res Hepatol. Gastroenterol. 44(6), 885–893 (2020).
- 7. . Epidemiology of gallbladder cancer. Clin. Exp. Hepatol. 5(2), 93–102 (2019).
- 8. . Biliary tract cancer: current medical treatment strategies. Cancers (Basel) 12(5), 1237 (2020).
- 9. . Biliary cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann. Oncol. 27, v28–v37 (2016).
- 10. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N. Engl. J. Med. 362(14), 1273–1281 (2010).
- 11. Second-line treatment in advanced biliary tract cancer: today and tomorrow. Anticancer Res. 40(6), 3013 (2020). •• Provides a summary of available targeted therapies that are currently in use or under investigation for previously treated advanced BTC patients.
- 12. . Second-line chemotherapy in advanced biliary cancer: a systematic review. Ann. Oncol. 25(12) 2328–2338 (2014). • Early systematic review and meta-analysis reporting pooled estimates of the efficacy of second-line treatments for advanced BTC patients, which concluded that there was insufficient evidence for recommending a second-line chemotherapy regimen in this patient population.
- 13. Nccn. Hepatobiliary Cancers, Version 5.2021, NCCN Clinical Practice Guidelines in Oncology. NCCN Clin. Pract. Guidelines Oncol. (2021).
- 14. Guidelines for the diagnosis and management of intrahepatic cholangiocarcinoma. J. Hepatol. 60(6), 1268–1289 (2014).
- 15. . Cholangiocarcinoma: a position paper by the Italian Society of Gastroenterology (SIGE), the Italian Association of Hospital Gastroenterology (AIGO), the Italian Association of Medical Oncology (AIOM) and the Italian Association of Oncological Radiotherapy (AIRO). Digest. Liver Dis. 42(12), 831–838 (2010).
- 16. . The current management of cholangiocarcinoma: a comparison of current guidelines. Biosci. Trends 10(2), 92–102 (2016).
- 17. Liver Cancer Study Group of Japan clinical practice guidelines for intrahepatic cholangiocarcinoma. Liver Cancer 11(4), 290–314 (2022).
- 18. Immunotherapy for biliary tract cancer in the era of precision medicine: current knowledge and future perspectives. Int. J. Mol. Sci. 23(2), 820 (2022). •• Describes the potential for immunotherapies to improve the prognosis of advanced BTC patients and provides an up-to-date summary of ongoing clinical trials evaluating the efficacy of immunotherapies in this patient population.
- 19. Second-line treatments in advanced biliary tract cancer: systematic literature review of efficacy, effectiveness and safety. Future Oncol. 18(18),2321–2338 (2022). • Reports that a lack of overlap among investigated treatment regimens precludes evaluation of the relative efficacy of particular regimens in second-line advanced BTC patients.
- 20. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372, n71 (2021).
- 21. . Single-arm trials in systematic reviews: are they adequately captured by existing search filters? Value Health 19(7), A358 (2016).
- 22. Cochrane Handbook for Systematic Reviews of Interventions. John Wiley & Sons (2019).
- 23. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. www.ohri.ca/programs/clinical_epidemiology/oxford.asp
- 24. Regorafenib after failure of gemcitabine and platinum-based chemotherapy for locally advanced/metastatic biliary tumors: REACHIN, a randomized, double-blind, phase II trial. Ann. Oncol. 31(9), 1169–1177 (2020).
- 25. A multi-institutional phase 2 trial of regorafenib in refractory advanced biliary tract cancer. Cancer 126(15), 3464–3470 (2020).
- 26. A phase 2 multi-institutional study of nivolumab for patients with advanced refractory biliary tract cancer. JAMA Oncol. 6(6), 888–894 (2020).
- 27. . Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med. Res. Methodol. 12(1), 9 (2012).
- 28. . Transformations related to the angular and the square root. Ann. Math. Stat. 607–611 (1950).
- 29. . The inverse of the Freeman–Tukey double arcsine transformation. Am. Statist. 32(4), 138–138 (1978).
- 30. . Meta-analysis in clinical trials. Control. Clin. Trials 7, 177–188 (1986).
- 31. . Chapter 10: Analysing data and undertaking meta-analyses. In: Cochrane Handbook for Systematic Reviews of Interventions version 6.3. Higgins JPTThomas JChandler J (Eds). Cochrane (2022).
- 32. . Meta-analysis of single-arm survival studies: a distribution-free approach for estimating summary survival curves with random effects. Stat. Med. 33, 2521–2537 (2014).
- 33. Hepatic arterial infusion with oxaliplatin and 5-FU/folinic acid for advanced biliary tract cancer: a phase II study. Dig. Dis. Sci. 58(8), 2399–2405 (2013).
- 34. Feasibility study of gemcitabine and cisplatin combination chemotherapy for patients with refractory biliary tract cancer. Invest. New Drugs 29(6), 1488–1493 (2011).
- 35. A phase II study of sunitinib as a second-line treatment in advanced biliary tract carcinoma: a multicentre, multinational study. Eur. J. Cancer 48(2), 196–201 (2012).
- 36. Ramucirumab plus pembrolizumab in patients with previously treated advanced or metastatic biliary tract cancer: nonrandomized, open-label, phase i trial (JVDF). Oncologist 23(12), 1407–e1136 (2018).
- 37. A multicenter phase II study of S-1 for gemcitabine-refractory biliary tract cancer. Cancer Chemother. Pharmacol. 71(5), 1141–1146 (2013).
- 38. Phase II study of fixed dose-rate gemcitabine plus S-1 as a second-line treatment for advanced biliary tract cancer. Cancer Chemother. Pharmacol. 80(6), 1189–1196 (2017).
- 39. Evaluation of combination nivolumab and ipilimumab immunotherapy in patients with advanced biliary tract cancers: subgroup analysis of a phase 2 nonrandomized clinical trial. JAMA Oncol. 6(9), 1405–1409 (2020).
- 40. Multicenter phase II study of S-1 monotherapy as second-line chemotherapy for advanced biliary tract cancer refractory to gemcitabine. Invest. New Drugs 30(2), 708–713 (2012).
- 41. Capecitabine with/without mitomycin C: results of a randomized phase II trial of second-line therapy in advanced biliary tract adenocarcinoma. Cancer Chemother. Pharmacol. 77(1), 109–114 (2016).
- 42. A phase 2 and biomarker study of cabozantinib in patients with advanced cholangiocarcinoma. Cancer 123(11), 1979–1988 (2017).
- 43. Single-institution experience with gemcitabine-cisplatin combination therapy as a second-line treatment for patients with unresectable biliary tract cancer after failure of gemcitabine-S-1 combination therapy: a prospective feasibility study. Cancer Chemother. Pharmacol. 81(5), 949–955 (2018).
- 44. A randomised phase II study of second-line XELIRI regimen versus irinotecan monotherapy in advanced biliary tract cancer patients progressed on gemcitabine and cisplatin. Br. J. Cancer 119(3), 291–295 (2018).
- 45. Efficacy and safety of FOLFIRINOX as salvage treatment in advanced biliary tract cancer: an open-label, single arm, phase 2 trial. Br. J. Cancer 122(5), 634–639 (2020).
- 46. . Efficacy and safety of capecitabine, irinotecan, gemcitabine, and bevacizumab as second-line treatment in advanced biliary tract cancer: a phase II study. Oncology 94(1), 19–24 (2018).
- 47. Capecitabine plus oxaliplatin as a second-line therapy for advanced biliary tract cancers: a multicenter, open-label, phase II trial. J. Cancer 10(25), 6185–6190 (2019).
- 48. A phase 2 trial of regorafenib as a single agent in patients with chemotherapy-refractory, advanced, and metastatic biliary tract adenocarcinoma. Cancer 125(6), 902–909 (2019).
- 49. Enhanced antitumor effect of binimetinib in combination with capecitabine for biliary tract cancer patients with mutations in the RAS/RAF/MEK/ERK pathway: phase Ib study. Br. J. Cancer 121(4), 332–339 (2019).
- 50. A randomised phase II study of oxaliplatin/5-FU (mFOLFOX) versus irinotecan/5-FU (mFOLFIRI) chemotherapy in locally advanced or metastatic biliary tract cancer refractory to first-line gemcitabine/cisplatin chemotherapy. Eur. J. Cancer 154, 288–295 (2021).
- 51. Phase II trial of erlotinib and bevacizumab in patients with advanced upper gastrointestinal cancers. Acta Oncologica 51(2), 234–242 (2012).
- 52. The oral VEGF receptor tyrosine kinase inhibitor pazopanib in combination with the MEK inhibitor trametinib in advanced cholangiocarcinoma. Br. J. Cancer 116(11), 1402–1407 (2017).
- 53. Phase II study of mFOLFOX3 (5-fluorouracil, leucovorin, oxaliplatin) as second-line treatment after gemcitabine failure in patients with unresectable/metastatic biliary tract cancer. Cancer Chemother. Pharmacol. 75(4), 757–762 (2015).
- 54. Liposomal irinotecan plus fluorouracil and leucovorin versus fluorouracil and leucovorin for metastatic biliary tract cancer after progression on gemcitabine plus cisplatin (NIFTY): a multicentre, open-label, randomised, phase 2b study. Lancet Oncol 22(11), 1560–1572 (2021).
- 55. Efficacy of capecitabine plus irinotecan vs irinotecan monotherapy as second-line treatment in patients with advanced gallbladder cancer: a multicenter phase 2 randomized clinical trial (GB-SELECT). JAMA Oncol 7(3), 436–439 (2021).
- 56. Phase II trial of trifluridine/tipiracil in patients with advanced, refractory biliary tract carcinoma. Oncologist 25(5), 380–e763 (2020).
- 57. Updated results of a phase IIa study to evaluate the clinical efficacy and safety of erdafitinib in Asian advanced cholangiocarcinoma (CCA) patients with FGFR alterations. J. Clin. Oncol. 37(Suppl. 15), 4117–4117 (2019).
- 58. A phase I pharmacokinetic and safety study of paclitaxel injection concentrate for nano-dispersion (PICN) alone and in combination with carboplatin in patients with advanced solid malignancies and biliary tract cancers. Cancer Chemother. Pharmacol. 87(6), 779–788 (2021).
- 59. Phase 2 study of lenvatinib monotherapy as second-line treatment in unresectable biliary tract cancer: primary analysis results. BMC Cancer 20(1), 1105 (2020).
- 60. Phase II study of BGJ398 in patients with FGFR-altered advanced cholangiocarcinoma. J. Clin. Oncol. 36(3), 276–282 (2018).
- 61. Second-line FOLFOX chemotherapy versus active symptom control for advanced biliary tract cancer (ABC-06): a phase 3, open-label, randomised, controlled trial. Lancet Oncol. 22(5), 690–701 (2021).
- 62. Multicenter phase II trial of axitinib monotherapy for gemcitabine-based chemotherapy refractory advanced biliary tract cancer (AX-BC study). Oncologist 26(2), 97–e201 (2021).
- 63. . Immunotherapy in advanced biliary tract cancers. Cancers (Basel) 13(7), 1569 (2021).
- 64. Lenvatinib plus pembrolizumab for patients with previously treated biliary tract cancers in the multicohort phase II LEAP-005 study. J. Clin. Oncol. 39(Suppl. 3), 321–321 (2021).
- 65. Efficacy and safety of trametinib in Japanese patients with advanced biliary tract cancers refractory to gemcitabine. Cancer Sci. 109(1), 215–224 (2018).
- 66. Phase II study of hepatic arterial infusion chemotherapy with oxaliplatin and 5-fluorouracil for advanced perihilar cholangiocarcinoma. Asia Pac. J. Clin. Oncol. 28, 28 (2021).
- 67. Sorafenib in patients with advanced biliary tract carcinoma: a phase II trial. Br. J. Cancer 102(1), 68–72 (2010).
- 68. Efficacy and safety of pembrolizumab for the treatment of advanced biliary cancer: Results from the KEYNOTE-158 and KEYNOTE-028 studies. Int. J. Cancer 147(8), 2190–2198 (2020).
- 69. . Second-line therapies in advanced biliary tract cancers. Lancet Oncol. 21(1), e29–e41 (2020). • Reviews available and emerging second-line treatment options for patients with advanced BTC.
