Biliary Tract Cancer: Current Medical Treatment Strategies

Background: Biliary tract cancers (BTCs) include cholangiocarcinomas and gallbladder cancers usually present at an advanced stage, which are considered resectable in less than 20% of cases and characterised by poor prognosis. Methods: In this review, we discussed the most recent therapeutic options on the basis of the most updated and complete reviews and recent prospective studies in selected BTC patients. Results: Due to the high recurrence rate of BTCs, we suggest the new recommendations that have been made on adjuvant chemotherapy and radiotherapy treatment after surgery. New chemotherapy combinations in advanced-stage patients allow a better survival benefit than the standard treatment. Furthermore, the revelation of complex molecular events and their interactions and relationships with some risk factors allowed the development of targeted/toxic agents alone or combination with chemotherapy that is really promising. In unresectable patients, hepatic arterial infusion of high-dose chemotherapy or selective internal radiotherapy could offer a primary mass volume reduction or its resection with the maintenance of liver function. Conclusions: The therapeutic landscape for BTCs is blooming again, the knowledge of their biology is still growing, but the available data on chemotherapy, radiotherapy, locoregional treatments, and target therapies have added hopes to improve patient survival.


Introduction
Tumours arising from the epithelium along the biliary tree are referred to as biliary tract cancers (BTCs), which include cholangiocarcinoma (CCA) and gallbladder cancer. Cholangiocarcinomas (CCAs) can then be differentiated by their anatomical origin into intrahepatic CCA (iCCA) and extrahepatic CCA (eCCA). Lesions involving the second-order bile ducts are defined as iCCA and accounts for 5-10% of all CCAs. In the other hand, eCCAs are divided into perihilar CCAs (pCCAs), when they lie above the cystic duct (CD), accounting for 60-70% of eCCAs, and distal CCAs (dCCAs), when they lie below the CD, accounting for the remaining part [1].
Gallbladder cancer is the most frequent biliary tract cancer, comprising 80-90% according to autopsy studies [2,3], followed by CCAs, which also represents the second most common primary liver cancer accounting for nearly 15% of all liver malignancies [4] and for approximately 3% of all gastrointestinal tumours [5]. The American Cancer Society's estimates a total of 11,980 newly diagnosed cases for approximately 4090 deaths from gallbladder and bile duct cancers in the United States for 2020 [6].
A meta-analysis from Messina et al. [19] has shown that adjuvant CT provides a mild benefit in terms of RFS (HR: 0.83, 95% CI 0.69-0.99) and effect on OS, but not in the overall population (HR: 0.91, 95% CI 0.75-1.09) nor the node positive subgroup (HR: 0.84, 95% CI 0.65-1.08), despite a higher risk of ≥ G3-G4 adverse events (AEs; RR: 3.03, 95% CI 2. 22-4.15). In contrast, Caparica et al. [20], analyzing those three clinical trials, stated that no difference in OS was observed between chemotherapy and observation, whereas a significant improvement was seen in RFS. Furthermore, a trend in benefit from adjuvant chemotherapy was observed in N+ patients. However, the lack of individual patient data, long recruitment periods, differences in clinical trial design, heterogeneity in CT regimens and patient's characteristics, different access to chemotherapy delivered after disease recurrence, and no stratification of patients according to BTC subtype could have skewed the results [19]. Lamarca et al. [21] found other limitations in the design of these studies, such as follow-up time, power, and data maturity.
Based on these findings, the ASCO guidelines have [7] recommended adjuvant capecitabine for a period of six months following curative resection of BTCs (CCA and GBC). Researchers agree on the fact that 12-16 weeks of recovery time is required after surgery prior to starting adjuvant therapy, and the administration of chemotherapy during a limited period of time (rather than until disease recurrence) is accepted as standard practice in the adjuvant setting and has been shown to be safe. Finally, based on the data available, chemotherapy based on fluoropyrimidines seems to be preferable [22]. This was also confirmed by the Japanese Phase II randomized KHBO 1208 trial [16] that compared 6-months of CT with biweekly Gem to tegafur/gimeracil/oteracil (S-1) in BTC Asian patients after major hepatectomy. S-1 has shown superior outcomes (p < 0.10) in terms of OS and 2-year RFS. Nowadays, the Phase III Japanese ASCOT trial is testing S-1 vs. observation in patients with resected BTC [17], while the multinational ACTICCA-01 trial [18] is testing cisplatin plus gemcitabine (the CisGem regimen), which is the current standard of care in patients with advanced disease and in patients with curatively resectable disease (Table 1 shows their regimen schedules). Moreover, the trial also plans to test adjuvant radiotherapy after R1 resection (and no disease progression after adjuvant chemotherapy).

Neoadjuvant Therapy
Neoadjuvant therapy could allow us to treat more patients than adjuvant therapy; this suggestion is mainly based on the observation that only 55% of patients who received capecitabine in the BILCAP trial have completed the planned eight cycles of treatment. However, neoadjuvant chemotherapy administration is not always possible due to the patient symptoms such as jaundice or other morbidities that could hold back treatment administration [23]. Unfortunately, no studies have been published on this specific area.

Advanced Disease: First-Line Treatment
The milestone of first-line treatment for BTC comes from the ABC-02 study [12] that suggested the efficacy of gemcitabine-cisplatin (GemCis) chemotherapy. Although this regimen has remained the standard of care, the median overall survival is still reported at just under 1 year [24]. Shroff et al. [25] demonstrated that the addition of nab-paclitaxel to standard doublet therapy (known as the GAP regimen: gemcitabine, nab-paclitaxel, and cisplatin) has improved survival. Notably, owing to poor tolerability observed, Gem and nab-paclitaxel doses were reduced by 20% from the standard dose. Interestingly, 12 patients were converted to resectable disease and completed surgery, 2 of whom achieved a pathologically complete response (PCR). Only 16% of patients withdrew owing to adverse effects, despite the high rate (58%) of grade 3 or higher adverse events [24,25]. The reduced-dose triplet is being evaluated against GemCis in the ongoing Phase III SWOG 1815 trial (NCT03768414).
The superiority of the combination regimen of folinic acid, fluoropyrimidine, irinotecan, and oxaliplatin (FOLFIRINOX) over Gem in pancreatic cancer patients [26] has led to the rationale for using this regimen in BTC patients. In a small retrospective series, FOLFIRINOX, as first-line treatment for BTC, has led to a disease control rate of 75% and OS of 15 months [27,28]. The ongoing Phase II/III PRODIGE38-AMEBICA trial (NCT02591030) aims to demonstrate an improvement in OS of 4 months in favour of the modified FOLFIRINOX (no 5-FU bolus on day 1) vs. GemCis [29]. The NIFE Phase II trial aims to challenge the current palliative first-line therapy for BTC by the use of nanoliposomalirinotecan/5-FU/leucovorin (nal-IRI). This trial is based on the assumption that ≥60% of patients will be progression-free after 4 months of nal-IRI [30]. In Table 2, first-line treatments are compared.

Advanced Disease: Beyond the First Line
After first-line therapy, 25%-50% of patients who progressed were still able to receive second-line chemotherapy (CT2) [31]. Discordant opinions have been suggested regarding the potential benefit of CT2 with a marginal activity (RR 7.2%, DCR 49.5%), and limited efficacy in unselected populations (OS 7.2 months, PFS 3.2 months). The ABC-06 Phase III trial demonstrated OS benefit with the modified combination of folinic acid, fluorouracil and oxaliplatin (mFOLFOX6) plus active symptom control (ASC) vs. the effect of ASC alone, in patients who had progressed to first-line GemCis. Notably, the effect of mFOLFOX6 was maintained regardless of previous platinum sensitivity, and 20% of patients experienced disease control of at least 9 months [21]. However, the study did not compare mFOLFOX6 regimen with a fluoropyrimidine alone arm, so it was not clear if the combination therapy was more efficient than monotherapy in this setting [32]. CT2 with the combination of mitomycin C and capecitabine vs. capecitabine alone yielded poor outcomes; oxaliplatin plus irinotecan (XELIRI) achieved PFS and OS benefit over irinotecan monotherapy [33] in a Chinese population the combination of folinic acid, fluorouracil and irinotecan (FOLFIRI) plus bevacizumab [34] and the sequence Gemox-FOLFIRI achieved 21.9 months OS in a small patient series [35]. In addition, trifluridine/tipiracil (FTD/TPI) was tested in BTC patients in II-line; as expect, no objective response (OR) was seen [36], although 60% of patients had been treated with up to three chemotherapy lines. A Phase II single-arm trial evaluating the combination of FTD/TPI and irinotecan for patients with refractory BTC is ongoing (NCT04072445), supporting the idea that this combination will likely have a higher response rate. The meta-analysis from Ying and Chen [37] has indicated that 1-year OS was 21.1% for a single targeted agent, 29.6% for a single toxic agent (mainly fluoropyrimidine alone), 26.9% for gemcitabine-based combination therapy, and 23.3% for taxanes-based combination therapy; in fact, in all cases, it was higher than the 15% observed in fluoropyrimidine-based combination therapy. However, no head-to-head comparison data were available for combination therapy vs. monotherapy [37].
These studies (summarised in Table 3) have shown heterogeneous results which emphasises the need of an appropriate selection of patients. Many single institutional studies have tried to define some prognostic tools for selecting patients who could benefit from CT2, such as age, performance status, surgery for primary tumour, a good response to first-line therapy, PFS longer than 6 months, neutrophil-to-lymphocyte ratio, and carcinoembryonic antigen, CA19-9, leucocyte, cholinesterase and albumin concentrations [32]. Only one multicentre European study has developed and validated (in three countries) a prognostic model based on four independent prognostic factors: the baseline performance status, the reason of ending the first line-therapy, surgical treatment for the primary tumour, and the presence of peritoneal carcinomatosis [32].

Radiotherapy
As stated by European Society of Medical Oncology(ESMO) guidelines, radiotherapy may be considered for certain patients, but there is limited clinical evidence for its effectiveness; therefore they are not currently commonly used in Europe outside of clinical trials. In particular, radiotherapy can be used to relieve some symptoms of biliary tract cancer-like pain and other symptoms by shrinking tumours that block blood vessels or bile ducts, or press on nerves [38]. A retrospective analysis of inoperable iCCA patients treated with definitive radiation therapy suggested that a biologically effective dose (BED) >80.5 Gy seems to be ablative, with long-term survival rates that compare favorably to resection [39]. The metanalysis of Ren et al. showed a 5-year OS benefit for eCCA and gallbladder cancer patients treated with adjuvant radiotherapy than without, with a major benefit in for those with lymph node-positive disease (OR = 0.15; 95% CI 0.07-0.35; p < 0.00001) and margin-positive disease (OR = 0.40; 95% CI 0.19-0.85; p = 0.02) [40]. Manterola et al. [41] have completed a literature review on the combination of chemo-radiotherapy (CRT) with fluoropyrimidines as a radio-sensitizer in the adjuvant setting or radiotherapy (RT) alone for resected gallbladder cancer. The effectiveness of adjuvant CRT continues to be controversial, although, in the SWOG080 trial, adjuvant CT with gemcitabine plus capecitabine (Gem-cape) followed by concomitant chemoradiation resulted in an impressive 2-year OS rate of 65% (95% CI, 53%-74%) among 79 patients with resected BTC, independently of the resection margin status [42]. Unfortunately, the paucity and low quality of data, including the lack of direct comparisons, did not permit a consensus on the modality of radiation, dose, or the adequate time for administration. It could reasonably be assumed that CRT could be more beneficial on OS than other adjuvant treatments with less-toxic effects due to the low radiation doses used [42].
However, based on these findings the ASCO guidelines have [7], the role of chemo-radiotherapy remains unclear and this suggests that its use should be limited to patients with eCCAs with R1 resection or other high-risk factors.

Targeted Therapy
Early studies on the genetic pathophysiology of CCAs focused on the elucidation of the carcinogenic role of single genes, often found mutated in other malignancies, or genes affected by copy number alterations [43]. In recent years, due to the advent of next-generation sequencing technology, a series of complex molecular events, their interactions, and their relationships with risk factors and causative events were revealed in biliary cancers. Among them, mutations in the genes of isocitrate dehydrogenases (IDH, isoforms 1 and 2), fusions of the fibroblast growth factor receptor 2 (FGFR2), and mutations of chromatin-remodeling genes, such as AT-rich interaction domain 1A (ARID1A), protein poly-bromo1 (PBRM1), and BRCA1-associated protein 1 (BAP1), were detected [8]. Target treatments were extensively discussed in very recent reviews [8,31,43].
FGFR2 gene rearrangements code for constitutively activated fusion products, resulting in aberrant mitogenic signalling with dysregulation of cell proliferation, survival, and angiogenesis. FGFR2 alterations have been reported in 13% to 17% of patients with CCAs [31,43]. Multiple oral tyrosine kinase inhibitors (TKIs) targeting FGFR were tested in patients with advanced CCAs, who presented FGFR gene fusions or other alterations and progressed after prior first-line therapy. Infigratinib (BGJ398) [46], pemigatinib (INCB054828) [47], derazantinib (ARQ 087) [48], and TAS-120 [49] demonstrated ORRs of 31%, 36%, 21%, and 25%, respectively, with a median PFS of 6.8 months for infigratinib, 6.9 months for pemigatinib, 5.7 months for derazantinib (data from the study with TAS-120 are not yet mature), an OS of 12.5 months for infigratinib and 21.1 months for pemigatinib (for Derezantinib and TAS-120, OS is not yet reached), and a DCR of approximately 80%. The ongoing Phase III trials will compare some of these TKIs to GemCis for first-line treatment of advanced CCAs [31,43].
VEGF was found to be overexpressed in 53.8% of iCCAs and 59.2% of eCCAs [43]; different associations between bevacizumab and CT and multikinase inhibitors, such as vandetanib and regorafenib [52], were tested and have shown some activity; ramucirumab is also currently under investigation [8,31,43].
Activation of missense BRAF mutations is reported in 11%-22% of CCAs, depending on the studies' population [53,54]. Vemurafenib, in combination with an MEK inhibitor, appeared to be promising in cancers harboring BRAF V600 mutations [31]. Selumetinib, an inhibitor of MEK1/2 proteins, supported stable disease alone or in combination with chemotherapy. An ongoing Phase II trial is evaluating the combination of solumetinib with gemcitabine vs. gemcitabine alone. Furthermore, other promising MEK inhibitors, including refametinib, trametinib, and MEK162, are under investigation in several clinical trials [8].
Unsatisfactory results derived from the use of MET inhibitors such as cabozantinib (abnormal MET activation has been found in 12%-58% of iCCAs) PI3K/AKT/mTOR inhibitors showed exiguous responses despite 12.5% of gallbladder cancers displaying activated mutations of PIK3CA. The ongoing Phase II STARTRK-2 trial is investigating entrectinib in patients with solid tumours harboring NTRK, ROS1, or ALK gene rearrangements/fusions, including those with CCAs [55].
Hypermutated BTCs characterized by increased expression of PD1/PD-L1 have a poor prognosis. Less than 1% of patients had high microsatellite instability (MSI), and the trials with immune-checkpoint inhibitors showed infrequent responses in the microsatellite stable tumours [31]. Immunotherapy was tested in BTC patients with some partial response and disease control, with a 12-month OS of 33% [32].
Promising targets for potentially interesting inhibitors are the proteins of the JAK/STAT, Wnt/β-catenin, Hedgehog and Notch signalling pathways, inhibitors that target mutations in chromatin-remodelling genes, such as ARID1A, PBRM1, and BAP1, and inhibitors of cancer-associated fibroblasts [8]. New drugs, aimed at simultaneously targeting several cell types that are critically involved in the development and progression of CCA, are in development.
A curious observational study, published in JAMA Oncology, showed that the use of aspirin in patients with BTC was associated with a decreased risk of death (HR of 0.63 for gallbladder cancer and of 0.71 for CCA) which may be due to the drug's inhibition of platelet aggregation slowing the spread of metastatic cancer cells. The researchers noted that incidental aspirin users who had no prior history of using the drug had a greater survival benefit from postdiagnosis aspirin use than those who were prevalent users, but they did not have an explanation [63]. One hypothesis is that tumours that developed during aspirin consumption were less susceptible to its beneficial function.

Locoregional Treatments
Considering that most patients present with unresectable advanced disease confined to the liver at diagnosis, liver-directed therapy could offer disease control or, even better, a primary mass volume reduction with the maintenance of adequate liver function. The liver's dual blood supply preferentially delivers high doses of chemotherapeutic agents to the hepatic artery, which supplies nearly all of the tumour's blood flow, while blood delivered by the portal vein maintains the health of the non-neoplastic liver parenchyma. Because the liver clears the chemotherapeutic agent via first-pass metabolism, locoregional approaches could diminish the systemic toxic effects of chemotherapy. A single-institution, Phase II clinical trial [64] investigated hepatic arterial infusion (HAI) of high-dose chemotherapy combined with first-line systemic chemotherapy in patients with unresectable intrahepatic CCA. Via HAI, high-dose floxuridine, a precursor of fluorouracil chemotherapy, was delivered directly into the liver, while systemic GemOx chemotherapy was administered. A response rate of 58% and an excellent disease control rate of 84% in the primary tumour at 6 months was achieved. Four patients underwent resection, and 1 patient had a complete pathologic response. The median PFS was 11.8 months and the median OS was 25.0 months, with a 1-year OS rate of 89.5%. A response in the dominant liver mass helped to prevent liver-related complications, such as biliary obstruction and portal vein occlusion, which are often associated with interruption of systemic therapy. This finding suggests that longer survival was correlated with the response of the primary tumour. Interestingly the subgroup analysis showed significant improvement in survival in patients with IDH1/2 mutated tumours (2-year OS 90%, 95% CI, 73%-99%) vs. wild-type tumours (2-year OS 33%, 95% CI, 18%-63%) [64].
In another Phase II study, the yttrium-90 microspheres (MISPHEC) trial [65], the technique of radioembolization, also known as selective internal radiotherapy (SIRT), was used in locally advanced iCCA in combination with systemic therapy with the GemCis regimen. Radiolabeled microspheres were administered via the hepatic arteries, delivering radiotherapy when reaching the tumour vasculature. The results showed PFS of 14 months and a median OS of 22 months, with an increased number of downstaged tumours and chance of surgery [41]. In the SIRCCA phase II trial, SIRT followed by CIS-GEM chemotherapy vs. CIS-GEM chemotherapy alone as first-line treatment of patients with unresectable intrahepatic CCA is currently randomising patients with unresectable CCA to either chemotherapy alone or resin-microsphere SIRT followed by chemotherapy, and a study comparing SIRT with chemoembolization is ongoing [66].

Discussion
BTCs are infrequent cancers that have been poorly studied in the past decades and have few treatment options, low response rates, and bad prognosis. Their rarity and the difficulty of obtaining good diagnostic samples make the effective development of translational research in this field significantly difficult. One of the most important steps forward will be the definition of the true value of some of the risk factors (such as R1 ore node positivity), and to correctly use them as stratification factors to pre-plan a correct statistical analysis [22] and for tailored treatment strategies. Moreover, valid and specific biomarkers are needed to select patients who are more likely to benefit from specific adjuvant strategies. Last but not least, there is a need for in-depth translational research to understand the mechanisms of resistance and disease recurrence. In this regard, an appropriate study design is crucial, where new Phase III randomised studies, comparing experimental arms with an active (non-observational) control arm (currently: capecitabine), need to be initiated after careful thoughts.
Although complete surgical resection (R0) remains the most effective approach for prolonging survival, disease recurrence is not uncommon. Furthermore, even if surgery could be technically performed, this could not always be contemplated due to patient comorbidities. The development of interventional techniques with lower morbidity is therefore essential to prolong survival in that population. Locoregional therapies should offer an important next step to improve survival by controlling local disease progression. As a result, some groups are experimenting with modifications to photoactive compounds in PDT, probe modifications in RFA, and combinations of locoregional therapy and concurrent chemotherapy [4]. Some progress has also been made in first-line therapies. The triple therapy with gemcitabine-cisplatin plus nab-paclitaxel resulted in a higher median PFS, OS, and RR than the combination therapies reported previously [24]. Other combinations are under investigation and have shown promising results.
Moreover, after tumour progression, there is no clear evidence of whether a combination therapy can be superior to a single targeted/toxic agent. However, since the overall quantity and quality of data regarding second-line therapy is poor and considering the risk of bias in comparisons between the observation studies, prospective randomized studies are strongly recommended [37]. Due to the rarity of advanced BTC patients, future multi-institutional clinical trials should promote good-sized studies that focus on stratifying patients on the basis of their anatomic subtype and genetic drivers in order to predict response to the new therapeutic options and the prognosis.
Despite the considerable progress that has been made in molecularly profiling BTCs, knowledge on the cause-and-effect relationship between the molecular changes and transformation of normal biliary epithelium to invasive malignancy, and our understanding of the interactions among the various signalling pathways and components that produce the cancer cell phenotypes, are still lacking.
At the top of the list is the definition of some prognostic tools for selecting patients who could benefit more from adjuvant chemotherapy or radiotherapy or the combination of both. For example, we need some lines about which therapy is preferred based on the volume, extension of disease, the presence of jaundice or other symptoms, age, performance status, serum carcinoembryonic antigen and CA19-9 levels and liver function. In addition, for the selection of patients to be candidates for a first-line treatment, we must evaluate the recurrence time, the primary surgery, the extent and locations of the metastatic disease and whether it may be susceptible to locoregional therapies, comorbidities and life expectancies. Unfortunately, no studies have yet established clear selection criteria and we hope this will be the topic of the next work.

Conclusions
The therapeutic landscape in BTC has expanded considerably in recent years after being nearly forgotten for a few decades. Combined therapy with chemotherapy, locoregional treatment and immunotherapy are promising strategies. The knowledge of the biology of BTC is still limited compared to that of other solid cancers, but the available data on target therapies have added hopes for the future management of hepatobiliary cancers and will likely continue to improve patient outcomes. Results from ongoing clinical trials are eagerly awaited to further elucidate the optimal management of this aggressive malignancy.