Abstract
Background:
Most patients with localized cholangiocarcinoma (CCA) endure cancer relapse after curative resection underscoring the importance of systemic therapy. The current study attempts to determine the impact of perioperative chemotherapy (PC) on survival in patients with CCA undergoing resection.
Methods:
Patients diagnosed with CCA undergoing curative-intent resection between January 1, 2000, and December 31, 2019, in a tertiary care center were included. Cox proportional hazard modeling was used to determine the impact of PC on disease-free survival (DFS) and overall survival (OS). In addition, a nomogram was constructed to estimate 3-year DFS.
Results:
Among the 182 patients included in the analysis, 102 underwent surgery alone, and 80 received surgery plus PC. Forty-two patients received neoadjuvant therapy, and 38 patients received adjuvant therapy. On multivariate analysis, PC was significantly associated with an improved DFS (HR, 95% CI: 0.63, 0.41–0.98; p = 0.04) and OS (HR, 95% CI: 0.46, 0.27–0.78; p < 0.01). In the interaction analysis, the survival benefit was especially seen in patients with positive resection margins and tumor size > 5 cm.
Conclusion:
In patients with CCA undergoing curative resection, receipt of PC was associated with improved DFS and OS. The nomogram constructed from this database provides an estimate of 3-year DFS after surgical resection. Randomized trials are needed to define the optimal regimen and sequence.
Keywords: Cholangiocarcinoma, Biliary tract cancer, Perioperative chemotherapy, Adjuvant chemotherapy
1. Introduction
Cholangiocarcinoma (CCA) encompasses a heterogeneous group of tumors originating from the epithelial lining of the bile ducts [1]. Based on the anatomical location of the tumor origin, CCA is broadly divided into three major subcategories: intrahepatic (iCCA), perihilar (pCCA), and distal (dCCA) CCA [2]. CCA is the second most common primary malignancy of the liver, with 23,000 new cases diagnosed annually in the United States (US) [3]. Of note, the incidence of CCA is increasing in the US [4] and globally [5], urging further research in this area. CCA is characterized by aggressive biology, and over 65% of patients present with an advanced-stage disease precluding curative-intent surgery, explaining the poor 5-year survival rate of 7–20% [6]. Approximately one-third of patients with CCA present with localized or locally advanced disease amenable to surgical resection [1]. However, despite resection, cancer recurrence is observed in a large proportion of patients, underscoring the importance of systemic therapy that could potentially eliminate micrometastatic disease and impact survival. The BILCAP study [7] has established adjuvant chemotherapy (AC) with capecitabine as the standard of care in patients with resected CCA, primarily based on improved survival observed in the adjuvant capecitabine arm in the prespecified per-protocol analysis. However, the BILCAP study did not show a statistically significant survival benefit with adjuvant capecitabine in the intention-to-treat analysis. In addition, the BILCAP study only included patients who underwent upfront surgical resection. In real-world practice, patients with CCA often present with a borderline resectable disease requiring neoadjuvant chemotherapy (NAC) to shrink the tumor and facilitate surgical resection with negative margins. Furthermore, many patients are unable to receive AC after resection for various reasons including post-surgical complications. The overall impact of perioperative chemotherapy (PC), that includes neoadjuvant and adjuvant chemotherapy, in resected CCA patients is not clearly defined. The current study aimed to address this knowledge gap and evaluated the impact of PC in patients with CCA who underwent curative-intent resection in a tertiary care center.
2. Methods
2.1. Study population
The study population consisted of patients with pathologically confirmed CCA undergoing curative-intent resection at Mayo Clinic, Rochester, MN, between January 1, 2000, and December 31, 2019. The institutional tumor registry was queried to identify the study population. Electronic medical records of the patients were reviewed to gather data on patient and tumor characteristics, detailed treatment, surgery, pathology, and survival. Specifically, the data on patient characteristics included patient demographics, Eastern Cooperative Oncology Group (ECOG) performance status, presenting symptoms, and co-morbidities. Tumor characteristics included histological diagnosis, grades, TNM stage based on the American Joint Committee on Cancer (AJCC) TNM Staging system (8th edition) [8], and resection margin status. The Institutional Review Board approved the study and deemed it not to require informed consent.
2.2. Statistical analysis
We summarized categorical data as frequency counts and percentages and continuous measures as means, standard deviations, medians, and ranges. Categorical variables were compared using the chi-square test or Fisher’s exact test. Continuous variables were compared using the one-way ANOVA test or Kruskal–Wallis test. Analyses of overall survival (OS) and disease-free survival (DFS) were conducted on the overall analysis population. OS was defined as the time from the date of diagnosis to death. Patients were censored at the time of the last follow-up. DFS was defined as the time from the date of diagnosis to recurrence or death, whichever happened first. Patients were censored at the time of the last follow-up. The distributions of OS and DFS were estimated using the Kaplan-Meier method and compared using log-rank tests. Univariate and multivariate analyses were conducted using Cox proportional hazards models. Additional analyses to determine factors predictive of the efficacy of PC were performed using Cox proportional hazards models with interactions of treatment and risk factors. A nomogram was constructed based on the final Cox model of DFS. Additionally, sensitivity analysis using propensity score (PS) methods, including both PS stratification and inverse probability of treatment weighting (IPTW) methods, was conducted to evaluate the treatment efficacy of PC. All analyses were conducted using two-sided tests with a significance level of 0.05.
3. Results
3.1. Patient and tumor characteristics
A total of 182 underwent surgical resection; surgery alone in 102 patients, and 80 patients received PC in addition to surgery (NAC in 42 and AC in 38 patients). Out of 42 patients receiving NAC, 20 patients also received AC. Gemcitabine plus cisplatin (70%) was most common NAC regimen used. Other regimens included gemcitabine plus cisplatin plus nab-paclitaxel, gemcitabine plus oxaliplatin and gemcitabine plus carboplatin. Amongst patients receiving AC, single agent capecitabine (39%) and gemcitabine plus cisplatin (37%) were most frequently administered regimens. Other therapies included gemcitabine plus capecitabine and capecitabine plus oxaliplatin. The median duration of treatment was 5.1 months. Table 1 summarizes the patient and tumor characteristics. The median (95% CI) follow-up for OS from surgery was 3.6 (2.9, 4.6) years, and from diagnosis 3.8 (3.2, 4.9) years. Distribution by gender was similar, with 99 (54.4%) patients being males. Almost all patients in this study cohort had an excellent (97.2%) ECOG performance status of either 0 or 1. Regional lymph node metastases were reported in 53 patients (29%) at the time of diagnosis. The cohort consisted of 65 (35.7%) patients diagnosed at a locally advanced stage (III/IV), while 117 patients (64.3%) were diagnosed with the earlier-stage disease (stages I or II). R0 resection or negative surgical resection margins was achieved in 162 patients (89%).
Table 1.
Patient and tumor characteristics.
| Surgery plus neoadjuvant/adjuvant therapy (N = 80) | Surgery alone (N = 102) | Total (N = 182) | p-value | |
|---|---|---|---|---|
| Gender | 0.6564a | |||
| Female | 35 (43.8%) | 48 (47.1%) | 83 (45.6%) | |
| Male | 45 (56.3%) | 54 (52.9%) | 99 (54.4%) | |
| Age at diagnosis | <0.0001b | |||
| Median | 62.0 | 68.7 | 65.5 | |
| Range | (18.7–83.0) | (39.8–90.3) | (18.7–90.3) | |
| Race | 0.5713a | |||
| White | 66 (82.5%) | 79 (77.5%) | 145 (79.7%) | |
| African American | 1 (1.3%) | 4 (3.9%) | 5 (2.7%) | |
| Asian | 1 (1.3%) | 1 (1.0%) | 2 (1.1%) | |
| American Indian | 0 (0.0%) | 2 (2.0%) | 2 (1.1%) | |
| Unknown or refused | 12 (15.0%) | 16 (15.7%) | 28 (15.4%) | |
| BMI | 0.4117b | |||
| Median Range | 27.7 (15.0–112.1) | 27.4 (17.6–46.7) | 27.6 (15.0–112.1) | |
| ECOG | 0.0153a | |||
| 0 | 61 (76.3%) | 57 (55.9%) | 118 (64.8%) | |
| 1 | 18 (22.5%) | 41 (40.2%) | 59 (32.4%) | |
| 2 | 1 (1.3%) | 4 (3.9%) | 5 (2.7%) | |
| Location | 0.9671a | |||
| Intrahepatic | 52 (65.0%) | 66 (64.7%) | 118 (64.8%) | |
| Extrahepatic | 28 (35.0%) | 36 (35.3%) | 64 (35.2%) | |
| T-Stage | 0.0358a | |||
| T1 | 30 (37.5%) | 48 (47.1%) | 78 (42.9%) | |
| T2 | 35 (43.8%) | 46 (45.1%) | 81 (44.5%) | |
| T3 | 7 (8.8%) | 7 (6.9%) | 14 (7.7%) | |
| T4 | 8 (10.0%) | 1 (1.0%) | 9 (4.9%) | |
| N-Stage | 0.0027a | |||
| N0 | 46 (57.5%) | 82 (80.4%) | 128 (70.3%) | |
| N1 | 33 (41.3%) | 20 (19.6%) | 53 (29.1%) | |
| N2 | 1 (1.3%) | 0 (0.0%) | 1 (0.5%) | |
| Stage | 0.0016a | |||
| I | 20 (25.0%) | 43 (42.2%) | 63 (34.6%) | |
| II | 19 (23.8%) | 35 (34.3%) | 54 (29.7%) | |
| III | 15 (18.8%) | 10 (9.8%) | 25 (13.7%) | |
| IV | 26 (32.5%) | 14 (13.7%) | 40 (22.0%) | |
| Grade | 0.0045a | |||
| G1 | 4 (5.0%) | 2 (2.0%) | 6 (3.3%) | |
| G2 | 48 (60.0%) | 38 (37.3%) | 86 (47.3%) | |
| G3 | 26 (32.5%) | 53 (52.0%) | 79 (43.4%) | |
| Aspirin use | 0.0299a | |||
| No | 44 (55.0%) | 72 (70.6%) | 116 (63.7%) | |
| Yes | 36 (45.0%) | 30 (29.4%) | 66 (36.3%) | |
| Time (days), Diagnosis to Surgery | <0.0001b | |||
| Median | 92.5 | 4.5 | 18.5 | |
| Time (days), Diagnosis to First therapy | <0.0001b | |||
| Median | 22.5 | 4.5 | 12.0 | |
| Hepatitis B | 0.2059a | |||
| No | 77 (96.3%) | 101 (99.0%) | 178 (97.8%) | |
| Yes | 3 (3.8%) | 1 (1.0%) | 4 (2.2%) | |
| Hepatitis C | 0.8055a | |||
| No | 78 (97.5%) | 100 (98.0%) | 178 (97.8%) | |
| Yes | 2 (2.5%) | 2 (2.0%) | 4 (2.2%) | |
| NASH | 0.2059a | |||
| No | 77 (96.3%) | 101 (99.0%) | 178 (97.8%) | |
| Yes | 3 (3.8%) | 1 (1.0%) | 4 (2.2%) | |
| PSC | 0.0084a | |||
| Missing | 1 | 0 | 1 | |
| No | 70 (88.6%) | 100 (98.0%) | 170 (93.9%) | |
| Yes | 9 (11.4%) | 2 (2.0%) | 11 (6.1%) | |
| Cholelithiasis | 0.1236a | |||
| Missing | 1 | 1 | 2 | |
| No | 67 (84.8%) | 93 (92.1%) | 160 (88.9%) | |
| Yes | 12 (15.2%) | 8 (7.9%) | 20 (11.1%) | |
| Diabetes | 0.7886a | |||
| No | 59 (73.8%) | 77 (75.5%) | 136 (74.7%) | |
| Yes | 21 (26.3%) | 25 (24.5%) | 46 (25.3%) | |
| Tumor Size | 0.0201a | |||
| 1–3 cm | 18 (22.5%) | 39 (38.2%) | 57 (31.3%) | |
| 3.1–5 cm | 16 (20.0%) | 30 (29.4%) | 46 (25.3%) | |
| 5.1–7 cm | 15 (18.8%) | 11 (10.8%) | 26 (14.3%) | |
| 7.1–10 cm | 21 (26.3%) | 15 (14.7%) | 36 (19.8%) | |
| >10.1 cm | 10 (12.5%) | 7 (6.9%) | 17 (9.3%) | |
| Histologic Type | 0.2031a | |||
| Missing | 0 | 2 | 2 | |
| Adenocarcinoma | 68 (85.0%) | 86 (86.0%) | 154 (85.6%) | |
| Adenosquamous | 2 (2.5%) | 0 (0.0%) | 2 (1.1%) | |
| Mixed HCC/Cholangiocarcinoma | 1 (1.3%) | 5 (5.0%) | 6 (3.3%) | |
| Other | 9 (11.3%) | 9 (9.0%) | 18 (10.0%) | |
| surgical resection margins | 0.0129a | |||
| Negative | 66 (82.5%) | 96 (94.1%) | 162 (89.0%) | |
| Positive | 14 (17.5%) | 6 (5.9%) | 20 (11.0%) | |
| Lymph nodes positive | 0.0128a | |||
| Missing | 7 | 18 | 25 | |
| No | 42 (57.5%) | 64 (76.2%) | 106 (67.5%) | |
| Yes | 31 (42.5%) | 20 (23.8%) | 51 (32.5%) |
Abbreviations: BMI, body mass index; ECOG, Eastern Co-Operative Group; NASH, non-alcoholic steatohepatitis; PSC, primary sclerosing cholangitis.
Chi-Square.
Kruskal Wallis.
There was no difference in the CCA location (i.e., intra-versus extrahepatic tumor) between patients who did or did not receive PC (p = 0.97). Only 33% of patients with stage I or II disease received PC compared to 63% of patients with stage III or IV disease. Table 1 describes the variables in both treatment groups. In summary, patients receiving PC were younger (median age 62 years vs. 68.7 years; p < 0.0001), had a superior ECOG PS (P = 0.01), higher T stage (p = 0.03), higher N stage (p = 0.0027), poorly differentiated tumors (p = 0.0045, and had higher prevalence of positive resection margin (p = 0.01).
Disease-Free Survival (DFS).
On univariate analysis (Supplement Table 1), inferior DFS was associated with the following factors: regional lymph node metastasis (p < 0.01), stage III/IV disease (p < 0.01), tumor size more than 5 cm in diameter (p = 0.03), and positive surgical resection margin (p = 0.02). The Median DFS of the patients who received PC and those who did not were 2.2 years (95% CI: 1.7–3.5) and 2.6 years (95% CI: 1.5–4.2), respectively (P = 0.91). If the DFS was calculated from date of diagnosis, median DFS was 2.5 years in patinets receiving PC and 1.7 years in patients not receiving PC.
In a multivariate Cox model, after adjusting for lymph node status, tumor stage, tumor size, tumor grade, and resection margin, receipt of PC was significantly associated with improvement in DFS (HR, 95% CI: 0.63, 0.41–0.98; p = 0.04) (Table 2). In addition, a positive surgical resection margin was also statistically significantly associated with poor DFS (HR, 95% CI: 1.89, 1.01–3.55; p = 0.04).
Table 2.
Multivariable analysis of disease-free survival (DFS).
| Multivariate Modeling - DFS | |||
|---|---|---|---|
| Variable | Comparison | HR (95% CI) | P- Value |
| Any Treatment | Neoadjuvant/Adjuvant Treatment vs. No Treatment | 0.632 (0.407, 0.983) | 0.0417 |
| N-Stage | N1 vs. N0 | 1.539 (0.770, 3.079) | 0.2226 |
| Stage | III/IV vs I/II | 1.488 (0.755, 2.935) | 0.2510 |
| Tumor Size | >5 cm vs≤5 cm | 1.413 (0.943, 2.118) | 0.0936 |
| Resection Margins | Positive vs. Negative | 1.890 (1.005, 3.552) | 0.0481 |
| Grade | G3/G4 vs G1/G2 | 0.947 (0.639, 1.404) | 0.7877 |
3.2. Overall-survival (OS)
On univariate analysis (supplement Table 2), we found that the factors that were associated with poor OS included the involvement of regional lymph nodes (p < 0.01), stage III/IV (p = 0.015), grade ≥3 histology (p = 0.04), tumor size more than 5 cm in diameter (p = 0.13), and positive surgical resection margins (p < 0.01). On univariate analysis, the median OS among the patients receiving PC was 5.7 years (95% CI: 3.5 - not estimable [NE]) compared to 4.3 years (95% CI: 2.8–6.9) in patients who did not receive PC. At 5 years, the OS rate in patients receiving PC was 56% (95% CI: 42.2–74.2) vs. 49.4% (39.7–61.6) among the patients who underwent surgery alone.
In the multivariate Cox model adjusted for tumor stage, tumor size, tumor grade, treatment status, N-Stage, and resection margins, receipt of PC was significantly associated with improved OS (Table 3). Patients who received PC had prolonged OS compared to the patients who did not receive PC (HR, 95% CI: 0.46, 0.28–0.78; p < 0.0041). Positive surgical resection margins and lymph node involvement were significantly associated with poor OS.
Table 3.
Multivariable analysis of overall survival (OS).
| Multivariate Modeling - OS | |||
|---|---|---|---|
| Variable | Comparison | HR (95% CI) | P- Value |
| Any Treatment | Neoadjuvant/Adjuvant Treatment vs. No Treatment | 0.464 (0.275, 0.783) | 0.0041 |
| N-Stage | N1 vs N0 | 2.790 (1.292, 6.022) | 0.0090 |
| Stage | III/IV vs I/II | 0.821 (0.394, 1.713) | 0.5996 |
| Tumor Size | >5 cm vs≤5 cm | 1.251 (0.783, 1.998) | 0.3487 |
| Resection Margins | Positive vs. Negative | 3.696 (1.834, 7.449) | 0.0003 |
| Grade | G3/G4 vs G1/G2 | 1.317 (0.833, 2.082) | 0.2381 |
3.3. Interaction Analysis:Disease-Free survival (DFS)
In interaction analyses (Fig. 1A), cancer stage (p-int<0.01), N-stage (p-int <0.01), resection margins (p-int <0.01), and tumor size (p-int = 0.04) were deemed predictive of efficacy of PC for DFS. Among the patients with stage III/IV CCA, patients receiving PC had longer DFS than those who did not (HR, 95% CI: 0.41, 0.24–0.73). Among patients with positive regional lymph nodes, patients receiving PC had longer DFS compared to those who did not receive PC (HR, 95% CI: 0.36, 0.19–0.68). In patients with positive surgical resection margins, administration of PC was associated with longer DFS compared to those who did not receive PC (HR, 95% CI: 0.18, 0.06–0.52). In patients who had tumor size more than 5 cm in diameter, PC was associated with prolonged DFS than those who did not receive PC (HR, 95% CI: 0.60, 0.36–1.02). Variables that did not predict DFS were tumor location, T-stage, and grade.
Fig. 1A. Interaction analysis: Disease-free survival (DFS).
Abbreviations: HR, hazard ratio; T-stage, tumor stage (size); N-stage, lymph node stage.
3.4. Overall survival (OS)
In interaction analyses (Fig. 1 B), tumor size (p-int <0.01) and resection margins (p-int = 0.01) were deemed predictive variables for OS. In patients with tumor size more than 5 cm, those who received PC had longer OS than those who did not (HR, 95% CI: 0.31, 0.16–0.61). In patients who had positive surgical resection margins, those who received PC had prolonged OS compared to those who did not receive PC (HR, 95% CI: 0.14, 0.43–0.45). The variables that did not predict OS were stage, tumor location, T-stage, lymph node involvement, and grade.
Fig. 1 B. Interaction analysis: Overall survival (OS).
Abbreviations: HR, hazard ratio; T-stage, tumor stage (size); N-stage, lymph node stage.
3.5. Nomogram
Based on the multivariable Cox proportional hazard regression model for DFS, administration of PC, stage, lymph node status, tumor size, and resection margins were selected to construct the nomogram. Based on the nomogram, a 3-year DFS rate in a patient with resected CCA could be predicted using the 5 variables. Positive resection margin carried the most weight, followed by receipt of PC. The total score ranged from 0 to 100 (Fig. 2).
Fig. 2. Nomogram predicting 3-year disease-free survival (DFS) in patients with resected cholangiocarcinoma.
Abbreviations: N-stage, lymph node stage; TNM, tumor-node-metastasis staging.
4. Discussion
The current study investigated the impact of systemic therapy administered in the perioperative period (i.e., PC) in a single-institution multicenter cohort of CCA patients undergoing curative-intent surgery. The study demonstrated statistically significant improvement in DFS and OS on multivariate analysis with the administration of PC in patients with resected CCA. In addition, several subgroups were identified who derived survival benefits from perioperative systemic therapy. A nomogram was constructed based on this study data that could predict DFS in resected CCA patients.
Surgical resection is the only known curative treatment modality for patients with CCA. However, over 65% of patients undergoing curative surgery endure cancer relapse resulting in poor survival in this group of patients [7,9]. An overwhelming amount of data suggest that the presence of micrometastatic disease or the minimal residual disease (MRD) after curative-intent surgery leads to cancer relapse [10]. Systemic therapy has been employed successfully in many tumor types to eliminate MRD, which led to long-term survival [11]. The sequence of systemic therapy administration varies based on the tumor type and stage of cancer-sometimes administered before surgery (neoadjuvant) and, more often, after the surgery (adjuvant). In the CCA treatment paradigm, systemic therapy has been studied in the post-operative (adjuvant) setting in the BILCAP study [7]. The BILCAP study randomized patients with resected CCA to either adjuvant capecitabine (n = 223) or observation (n = 224). After a median follow-up of 60 months, in the intention-to-treat analysis, the median overall survival was 51·1 months (95% CI, 34·6–59·1) in the capecitabine group vs. 36·4 months (95% CI, 29·7–44·5) in the observation group. The difference in the OS did not meet statistical significance (adjusted HR 0·81; 95% CI, 0·63–1·04; p = 0·097) in the intention-to-treat analysis. In the prespecified per-protocol analysis, however, the OS in the adjuvant capecitabine group was 53 months (95% CI, 40 to not reached) vs. 36 months (95% CI, 30–44) in the observation group, a difference which was statistically significant (adjusted HR 0·75; 95% CI, 0·58-0·97; p = 0·028). The BILCAP study result established adjuvant therapy with six months of capecitabine as the standard of care for patients with resected CCA. However, patients in the real-world setting often do not conform to the BILCAP trial criteria and often unable to receive adjuvant therapy because of various reasons, including surgical complications and delays in improvement in performance status. In a retrospective study based on the Surveillance Epidemiology and End Results (SEER) program database in patients with iCCA, only 41% of patients received AC [12]. As a result, we attempted to evaluate the survival impact of systemic therapy received in adjuvant or neoadjuvant settings on patients of CCA treated in a routine practice setting.
An overwhelming amount of data suggest that surgical resection of the tumor with a negative margin is a critical determinant of long-term survival [13]. CCA is often not resectable upfront and requires neoadjuvant therapy to downstage and make the tumor amenable to resection. Although the role of neoadjuvant therapy in patients with CCA is not established based on randomized trials, evidence exists suggesting that neoadjuvant therapy confers survival benefits [14,15]. A meta-analysis that included 1880 patients participating in 8 studies reported longer median OS in resected patients who received NAC versus those who did not (29 months vs. 12 months, P < 0.001) [14]. A propensity score-matched analysis based on the National Cancer Database reported longer OS in patients with CCA who received NAC compared to those who received AC following resection [15]. Hence, NAC is increasingly being utilized in clinical practice for patients with locally advanced CCA. As the primary goal of systemic therapy, both in the adjuvant and neoadjuvant setting, is the prolongation of survival, it is worthwhile to assess the overall impact of systemic therapy on survival irrespective of the timing of systemic therapy in relation to the surgery (adjuvant versus neoadjuvant). Recently, a phase II trial demonstrated the feasibility of neoadjuvant chemotherapy with gemcitabine, cisplatin and nab-paclitaxel for patients with resectable high-risk CCA [16].
As CCA is a heterogeneous entity, and the impact of PC likely varies considerably from one subgroup to another. In our analysis, PC did demonstrate an improvement in OS and DFS on multivariate analysis. We performed an analysis to identify patient subgroups deriving a significant survival benefit from PC. As locally advanced GI cancers (stage III or stage II tumors with large primary lesions) have a higher probability of harboring MRD after resection [17], systemic therapy often provides survival benefits to this group of patients, a finding firmly established in other cancers [18]. Our study result showed a similar conclusion that patients with higher T stage, lymph node involvement, and positive surgical resection margin derived survival benefit from PC. This study result further reinforces the argument supporting systemic therapy in locally advanced CCA.
Each clinicopathologic factor has a specific impact on survival in resected patients with CCA [19]. Consequently, it is critical to measure the impact of an individual clinicopathologic factor on survival. To that end, we constructed a nomogram to provide guidance on estimated survival in surgically resected patients. The nomogram based on receipt of PC, N-stage, overall TNM stage, tumor size, and the resection margin status gives an estimate of 3-year DFS probability. The information provided by the nomogram can help counsel patients in routine clinical practice and help select patients at a very high risk of cancer relapse for adjuvant clinical trials with novel molecules or strategies.
This study has several limitations that must be acknowledged while interpreting the findings. The patient cohort was heterogeneous, and patients received a variety of different treatment modalities over two decades, during which treatment procedures, including surgical techniques, have changed, which may have introduced biases. Additionally, the unmeasured confounding and selection biases inherent in a retrospective study could impact the results. However, we performed a multivariable analysis to determine the impact of PC in various patient subgroups to address this limitation. The current study also did not integrate the genomic biomarkers in the survival analysis, as genomic profiling was not available in the majority of patients. The modest sample size did not allow us to evaluate the impact of specific types of PC. Furthermore, the nomogram did not have a validation cohort, but such studies will be planned in the future.
In summary, the current study demonstrates that all patients with CCA undergoing curative-intent resection do not derive a survival benefit from PC uniformly. PC conferred a significant survival benefit to the patient subgroups with adverse prognostic features, specifical patients with lymph node involvement and larger tumors. These findings highlight the importance of detailed pathologic staging in the multimodality treatment of CCA patients. Prospective studies are needed to define clinicopathologic and molecular predictors of benefit from perioperative systemic therapies.
Supplementary Material
Footnotes
CRediT authorship contribution statement
Hind Hassan: Investigation, Writing – original draft. Sakti Chakrabarti: Writing – original draft. Tyler Zemla: Formal analysis, Writing – review & editing. Jun Yin: Formal analysis, Writing – review & editing. Vanessa Wookey: Investigation, Writing – review & editing. Kritika Prasai: Investigation, Writing – review & editing. Amro Abdellatief: Investigation, Writing – review & editing. Renuka Katta: Investigation, Writing – review & editing. Nguyen Tran: Writing – review & editing. Zhaohui Jin: Writing – review & editing. Sean Cleary: Writing – review & editing. Lewis Roberts: Writing – review & editing. Amit Mahipal: Conceptualization, Methodology, Supervision, Writing – review & editing.
Declaration of competing interest
None.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ejso.2023.106994.
References
- [1].Valle JW, Kelley RK, Nervi B, Oh D-Y, Zhu AX. Biliary tract cancer. Lancet 2021;397(10272):428–44. [DOI] [PubMed] [Google Scholar]
- [2].Banales JM, Marin JJG, Lamarca A, Rodrigues PM, Khan SA, Roberts LR, et al. Cholangiocarcinoma 2020: the next horizon in mechanisms and management. Nat Rev Gastroenterol Hepatol 2020;17(9):557–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA A Cancer J Clin 2020;70(1):7–30. [DOI] [PubMed] [Google Scholar]
- [4].Javle M, Lee S, Azad NS, Borad MJ, Kate Kelley R, Sivaraman S, et al. Temporal changes in cholangiocarcinoma incidence and mortality in the United States from 2001 to 2017. 2022. The oncologist. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Bertuccio P, Malvezzi M, Carioli G, Hashim D, Boffetta P, El-Serag HB, et al. Global trends in mortality from intrahepatic and extrahepatic cholangiocarcinoma. J Hepatol 2019;71(1):104–14. [DOI] [PubMed] [Google Scholar]
- [6].Alabraba E, Joshi H, Bird N, Griffin R, Sturgess R, Stern N, et al. Increased multimodality treatment options has improved survival for Hepatocellular carcinoma but poor survival for biliary tract cancers remains unchanged. Eur J Surg Oncol 2019;45(9):1660–7. [DOI] [PubMed] [Google Scholar]
- [7].Primrose JN, Fox RP, Palmer DH, Malik HZ, Prasad R, Mirza D, et al. Capecitabine compared with observation in resected biliary tract cancer (BILCAP): a randomised, controlled, multicentre, phase 3 study. Lancet Oncol 2019;20(5):663–73. [DOI] [PubMed] [Google Scholar]
- [8].Amin MB, Edge SB, Greene FL, Byrd DR, Brookland RK, Washington MK. In: et al. AJCC cancer staging manual. 8th ed. Chicago: AJCC; 2017. [Google Scholar]
- [9].Wang Y, Li J, Xia Y, Gong R, Wang K, Yan Z, et al. Prognostic nomogram for intrahepatic cholangiocarcinoma after partial hepatectomy. J Clin Oncol 2013;31(9):1188–95. [DOI] [PubMed] [Google Scholar]
- [10].Badia-Ramentol J, Linares J, Gomez-Llonin A, Calon A. Minimal residual disease, metastasis and immunity. Biomolecules 2021;11(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Luskin MR, Murakami MA, Manalis SR, Weinstock DM. Targeting minimal residual disease: a path to cure? Nat Rev Cancer 2018;18(4):255–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Altman AM, Kizy S, Marmor S, Hui JYC, Tuttle TM, Jensen EH, et al. Adjuvant chemotherapy for intrahepatic cholangiocarcinoma: approaching clinical practice consensus? Hepatobiliary Surg Nutr 2020;9(5):577–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Li MX, Bi XY, Li ZY, Huang Z, Han Y, Zhao JJ, et al. Impaction of surgical margin status on the survival outcome after surgical resection of intrahepatic cholangiocarcinoma: a systematic review and meta-analysis. J Surg Res 2016; 203(1):163–73. [DOI] [PubMed] [Google Scholar]
- [14].Kamarajah S, Giovinazzo F, Roberts KJ, Punia P, Sutcliffe RP, Marudanayagam R, et al. The role of down staging treatment in the management of locally advanced intrahepatic cholangiocarcinoma: review of literature and pooled analysis. Ann Hepatobiliary Pancreat Surg 2020;24(1):6–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Yadav S, Xie H, Bin-Riaz I, Sharma P, Durani U, Goyal G, et al. Neoadjuvant vs. adjuvant chemotherapy for cholangiocarcinoma: a propensity score matched analysis. Eur J Surg Oncol 2019;45(8):1432–8. [DOI] [PubMed] [Google Scholar]
- [16].Maithel SK, Javle MM, Mahipal A, Lin BS-L, Akce M, Switchenko JM, et al. NEO-GAP: a phase II single-arm prospective feasibility study of neoadjuvant gemcitabine/cisplatin/nab-paclitaxel for resectable high-risk intrahepatic cholangiocarcinoma. J Clin Oncol 2022;40(16_suppl):4097. [Google Scholar]
- [17].Chakrabarti S, Kasi AK, Parikh AR, Mahipal A. Finding waldo: the evolving paradigm of circulating tumor DNA (ctDNA)-Guided minimal residual disease (MRD) assessment in colorectal cancer (CRC). Cancers 2022,14(13). [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Andre T, de Gramont A, Vernerey D, Chibaudel B, Bonnetain F, Tijeras-Raballand A, et al. Adjuvant fluorouracil, leucovorin, and oxaliplatin in stage II to III colon cancer: updated 10-year survival and outcomes according to BRAF mutation and mismatch repair status of the MOSAIC study. J Clin Oncol 2015;33(35):4176–87. [DOI] [PubMed] [Google Scholar]
- [19].Yu T-h, Chen X, Zhang X-h, Zhang E-c, Sun C-x. Clinicopathological characteristics and prognostic factors for intrahepatic cholangiocarcinoma: a population-based study. Sci Rep 2021;11(1):3990. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.