The safety and efficacy of neoadjuvant chemotherapy based on our resectability criteria for locally advanced perihilar cholangiocarcinoma

doi:10.21203/rs.3.rs-2489659/v1

Download PDF

Research Article

The safety and efficacy of neoadjuvant chemotherapy based on our resectability criteria for locally advanced perihilar cholangiocarcinoma

https://doi.org/10.21203/rs.3.rs-2489659/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 Jul, 2023

Read the published version in Langenbeck's Archives of Surgery →

You are reading this latest preprint version

Background: Neoadjuvant chemotherapy (NAC) is not commonly used for perihilar cholangiocarcinoma (PHC). This study aimed to evaluate the safety and efficacy of NAC for PHC.

Patients and methods: Ninety-one PHC patients without metastases were treated at our department. They were classified as resectable (R), borderline resectable (BR), or locally advanced unresectable (LA). Upfront surgery (US) was performed for R-PHC without regional lymph node metastases (LNM). The NAC regimen consisted of two courses of gemcitabine-based chemotherapy for advanced PHC: R-PHC with LNM, BR, and LA.

Results: US and NAC were done on 32 and 59 patients, respectively. For US, 31 patients underwent resection (US-R); one with peritoneal-dissemination did not. NAC for locally advanced PHC was done in 56/59 (92%). NAC caused adverse effects in 10/59 (17%), allowed 36/59 (61%) to undergo resection (NAC-R) without impairing liver function, and spared 23/59 (39%) as un-resection (NAC-UR). Overall survival was better in the US-R and NAC-R groups than in NAC-UR (MST: 74 vs. 57 vs. 17 months, p<0.001). In 59 NAC patients, tumour size response occurred in 11/11 (100%) of R, 22/33 (66.7%) of BR, and 9/15 (60.0%) of LA patients. The un-resection rate was higher in the LA group (27% (3/11) in R, 30.3% (10/33) in BR, and 67% (10/15) in LA, p=0.039). Multivariate analyses revealed that LA and age were independent risk factors for un-resection after NAC.

Conclusion: NAC is safe, and contributes to curative resection for advanced PHC. LA remained a risk factor for non-resection through NAC, and required further interventions.

perihilar cholangiocarcinoma

neoadjuvant chemotherapy

upfront surgery

gemcitabine

resectability

Treatment of perihilar cholangiocarcinoma (PHC) remains among the most critical challenges in the modern era. Surgical resection with a histologically negative margin (R0) is the only potentially curative treatment [1–3]. However, this treatment is often impossible for distant metastases and local progressions, such as in cases where there has been invasion of the hepatic artery (HA) and portal vein (PV), or extensive bilateral spread to second-order biliary radicals [4–5]. Curative resection of locally advanced PHC generally requires major hepatectomy with concomitant resection and reconstruction of HA or PV, which is highly associated with morbidity and mortality [1, 2, 4, 6]. Furthermore, curative intent surgery for locally advanced PHC sometimes fails in non R0 resection, which is associated with early recurrence and poor prognosis [1, 2, 7, 8]. Therefore, it is necessary to allow advanced PHC to achieve R0 resection through intervention.

Some institutions have introduced neoadjuvant chemotherapy (NAC) for PHC, which is expected to manage micrometastatic diseases and reduce tumour sizes preoperatively, improving prognoses [9–11]. However, their studies had small cohorts and mainly included intrahepatic or distal cholangiocarcinomas. Significantly, the effect of NAC on liver function and the postoperative course has not been clarified. We categorised PHC into resectable (R), borderline resectable (BR), and locally advanced unresectable (LA) based on computed tomography (CT) images [11]. US was performed for R-PHC without regional lymph node (LN) metastases. NAC was conducted and curative intent resection was attempted for advanced PHC of R with LN metastases, BR, and LA. In this study, we aimed to evaluate our treatment strategies based on resectability criteria, highlighting the efficacy and safety of NAC for advanced PHC.

2.1. Patients

Between August 1st, 2010 and December 31st, 2020 109 patients were pathologically diagnosed with PHC at our department. After excluding 18 patients who received radiation therapy (n = 8), had distant metastasis (n = 3), impaired performance status (n = 5), chronic obstructive pulmonary disease (n = 1), or post-ERCP pancreatitis (n = 1), the remaining 91 patients with PHC were included in this study. The patients were treated according to our treatment flow, based on the resectability criteria described below. The Institutional Review Board of Mie University Hospital approved this retrospective study and waived the need for individual informed consent (No.2954).

2.2. Our resectability criteria

Our resectability criteria included biliary and vascular factors [10]. R was defined as curative resection on either side of the right or left trisectionectomy or less hepatectomy (biliary factor), where combined resection with reconstruction of the PV or HA was not needed (vascular factor). BR was defined as curative resection that could be obtained by either side of right or left trisectionectomy or less hepatectomy (biliary factor), and combined vascular resection with safe reconstruction of the PV or HA could be performed (vascular factor). LA was defined as curative resection not being obtained even by either side of the right or left trisectionectomy (biliary factor), or combined vascular resection with safe reconstruction of PV or HA could not be performed (vascular factor). Up-front surgery (US) was preferred for R patients when clinical evidence of LN metastasis was excluded based on positron emission tomography-computed tomography (PET-CT) and multidetector CT (MDCT) findings. NAC was conducted in the regional LN-positive R, BR, and LA patient groups, with the aim of curative intent surgery.

2.3. NAC

Gemcitabine-based chemotherapy was administered for NAC. Gemcitabine (800 mg/m2 on days 7 and 21) plus S-1 (GS; 80 mg/body daily on days 1–21 every four weeks), gemcitabine alone (1000 mg/m2 on days 1 and 15 every four weeks) (G), or gemcitabine (1000 mg/m2 on days 1 and 8) plus cisplatin (25 mg/m2 on days 1 and 8 every three weeks) (GC) was administered for two courses. A gemcitabine-based regimen was selected as NAC. When curative intent surgery was not indicated after two courses of NAC, chemotherapy was continued until the time of conversion to surgery.

2.4. Indication of curative intent surgery

The indication for curative resection was evaluated from the anatomical and liver function perspectives at the meetings of the attending multidisciplinary departments. Right hepatectomy was indicated for Bismuth types I, II, and IIIa, and left hepatectomy was indicated for Bismuth type IIIb. Ttrisectionectomy or central bisectionectomy is indicated for Bismuth type IV. Surgery was contraindicated in patients with apparent invasion of the umbilical portion (U portion) and the right posterior portal vein (P portion), and chemotherapy was continued. Combined vascular resection and reconstruction were performed for the local invasion of HA and PV to be able to reconstruct. However, vascular invasion beyond the second branch of the PV and HA is considered contraindicated because safe vascular reconstruction is generally impossible. In addition, liver function was evaluated using the indocyanine green plasma clearance rate (KICG) and the hepatic accumulation index 99mTc-GSA scintigraphy agent, which was calculated as the ratio of uptake by the liver alone to that by the liver plus heart at 15 min (LHL15) [12–14]. The future functional remnant liver ratio (f-rem) was calculated from the fusion image of MDCT and the hepatic uptake ratio of 99mTc-GSA scintigraphy at 15 min using 3D simulation software (Synapse Vincent; Fujifilm, Tokyo, Japan) [8]. Finally, the type of hepatectomy was determined based on the function of the future remnant liver (f-rem-KICG), which was calculated by multiplying f-rem by the KICG [8]. In a previous study, patients with f-rem-KICGs of less than 0.05 were not indicated for major hepatectomy [12]. Limited liver resection and extra bile duct resection were selected in cases of impaired liver function requiring major hepatectomy [15]. Portal vein embolisation was indicated if the estimated future remnant liver volume was < 40%, f-rem-KICG was < 0.05, or if right hepatectomy with pancreaticoduodenectomy was necessary.

2.5. Evaluation of perioperative clinical courses

The patients' clinical values were evaluated for the following: age, sex, UICC-TNM8th classification, resectability, Bismuth type, type of chemotherapy, white blood cells (/µL), neutrophils (/µL), lymphocytes (/µL), platelet count (/µL), albumin (g/dL), neutrophil-to-lymphocyte ratio (NLR), prognostic nutritional ratio (PNI), carcinoembryonic antigen (CEA, ng/mL), and carbohydrate antigen 19 − 9 (CA19-9, U/mL). CA19-9 levels were evaluated after biliary drainage. The other factors were assessed before US or NAC. Multiple department joint conferences have determined the indication for curative intent surgery based on anatomical resectability and future remnant liver function. In the patients with NAC, adverse events during NAC were categorized according to the National Cancer Institute's Common Toxicity Criteria for Adverse Events (CTCAE), version 5.0. Changes in laboratory findings including liver function were evaluated before and after NAC. Radiologists evaluated changes in tumour size using CT imaging studies, classifying them as decrease, no change, or increase. A decrease or no change was judged as a tumour size response. Surgical outcomes and pathological findings were assessed in the patients who underwent resection.

2.6. Adjuvant chemotherapy

From four to six weeks after resection, adjuvant chemotherapy (AC) was conducted for six months if tolerated by the patient. Between February 2005 and May 2013, the chemotherapy regimen consisted of gemcitabine at a dose of 800 mg/m2 biweekly. From June 2013, the regimen was S1 orally twice daily at 60 mg/m2/day on days 1–28 of a 42-day cycle.

2.7. Statistical analysis

Continuous and categorical variables were expressed as medians with ranges and compared using the Mann-Whitney and chi-square tests. In all patients who came for reassessment, the date of the initial treatment was chosen as the starting point for measuring the survival time. Comparisons of clinical factors before and after NAC were performed using Wilcoxon analysis. Overall survival was calculated using the Kaplan-Meier method, and comparisons between the groups were evaluated using the log-rank test. The day of the final follow-up was 31 December 31st 2021. All variables were dichotomised for the analysis. A multivariate analysis was performed using a multivariate hazard model. Variables with a significance of p < 0.25 in the univariate analysis were entered into the multivariate analysis. All statistical analyses were performed using SPSS version 27 software (SPSS Inc., Chicago, IL, USA). Statistical significance was set at p < 0.05.

The 91 PHC patients with PHC were treated according to our treatment flow (Fig. 1). Patients were classified into three groups according to the resectability criteria: R in 41 patients, BR in 35, and LA in 15. US was performed in 32 patients, 30 of whom were diagnosed with R without lymph node metastases, and the other two were diagnosed with NAC. Curative intent resection (US and curative intent resection: US-R) was performed in 31 patients, excluding one patient with peritoneal dissemination. NAC was conducted in 59 patients: R in 11, BR in 33, and LA in 15. Among the 11 patients with R, NAC was conducted for clinically regional LN metastases in seven patients and delayed surgery for prolonged jaundice or refractory cholangitis in the other four. Following curative intent, resection was performed on 36 patients (NAC and resection: NAC-R) but was not indicated in the other 23 patients (NAC and un-resection: NAC-UR) (Fig. 1). In the comparison of patient characteristics among US-R, NAC-R, and NAC-UR, PHC was significantly the most advanced in NAC-UR patients, as shown by the rates of cT4, cN1-2, LA, and CA19-9 levels. By contrast, PHC was significantly less advanced in the US-R group (Table 1).

Table 1

Patients’ background between US-R, NAC-R, and NAC-UR
	US-R n = 31	NAC-R n = 36	NAC-UR n = 23	p
Age, years	72 (44–87)	70 (50–84)	76 (61–81)	0.107
Sex, male/female	17/14	22/14	17/6	0.354
T4/<=T3	3/28	19/17	16/7	< 0.001
N0/1–2	3/28	15/21	10/13	0.006
Resectability, R/BR/LA	29/2/0	8/23/5	3/10/10	< 0.001
Bismuth type, I/II/IIIa/IIIb//IV	6/2/7/11/5	2/1/11/8/14	0/3/4/5/11	0.079
Types of NAC
GS/G/GC/GS to GC	0/0/0/0	31/1/2/2	20/1/1/1	< 0.001
Blood examination, initial value
White blood cell, /uL	4590 (3190–8010)	6035 (3060–11810)	5490 (3540–11000)	0.043
Neutrophil, /uL	2900 (1460–5860)	3555 (1420–9220)	3090 (1390–9320)	0.070
Lymphocyte, /uL	1300 (550–2510)	1565 (990–3040)	1480 (600–2440)	0.117
Platelet count, x10⁴ /uL	22.2 (11.7–43.8)	20.9 (10.2–51.1)	22.3 (11.6–45.7)	0.758
Albumin, g/dL	3.7 (2.6–4.6)	3.8 (3.0-5.1)	3.9 (2.6–4.6)	0.687
NLR	2.42 (0.97–8.15)	2.54 (1.19–4.78)	2.05 (0.77–10.6)	0.351
PNI	43.3 (31.2–55.6)	46.2 (38.1–61.5)	44.4 (35.2–56.2)	0.270
CEA, ng/mL	3.0 (0.7–38.4)	3.8 (0.9–21.5)	3.60 (1.1–11.9)	0.433
CA19-9, U/mL	58.7 (7.0-907)	116 (0.2-76452)	766 (5.2–9066)	0.001
US: upfront surgery, NAC: neoadjuvant chemotherapy, TNM8th, R: resectable, BR: borderline resectable, LA: locally advanced unresectable, G: gemcitabine, GS: gemcitabine plus S-1, GC: gemcitabine plus cisplatin, NLR: neutrophil-to-lymphocyte ratio, PNI: prognostic nutritional index, CEA: carcinoembryonic antigen, CA19-9: carbohydrate antigen 19 − 9. cT and cN were evaluated by UICC-TNM8th.

Table 2

Surgical outcomes of patients who underwent curative intent surgery
	US-R n = 31	NAC-R n = 36	P
Operative duration, min	609 (404–943)	602 (383–728)	0.692
Blood loss, mL	1367 (166–5166)	1749 (160–6059)	0.185
Preoperative portal vein embolization, yes/no	3/28	5/31	0.443
Operative procedures
S1,2,3,4	13	19	0.179
S1,2,3,4,5,8	1	2
S1,5,6,7,8	8	13
S1,4,5,6,7,8	0	0
S1,4,5,8	1	0
Other limited liver resection	3	0
Bile duct resection alone	5	2
Combined pancreatoduodenectomy, yes (%)	7/24 (22.6%)	1/35 (2.8%)	0.016
Combined PV resection and reconstruction, yes (%)	5 (16.1%)	18 (50.0%)	0.004
Combined HA resection and reconstruction, yes (%)	1 (3.23%)	8 (22.2%)	0.024
Pathological findings
G1/G2/G3	23/6/2	28/4/3*	0.653
pT0/1/2/3/4	0/2/19/7/3	1/1/19/7/8	0.517
pT, 4/1–3	3/28	8/28	0.167
pN0/1–2	17/14	19/17	0.866
pM0/1	30/1	33/3	0.366
R, 0/1/2	23/5/3	25/6/5	0.863
Morbidity and mortality
Complication (Clavien-Dindo ≧ III) (%)	15/16 (48.4%)	22/14 (61.1%)	0.296
Hospital stays, days	38 (22–146)	40 (13–161)	0.880
90-day mortality, yes	0	0	-
Adjuvant chemotherapy, yes/no	22/9	28/8	0.523
US: upfront surgery, NAC: neoadjuvant chemotherapy, S: segment, PV: portal vein, HA: hepatic artery, G: grade, R0: no residual tumour, R1: microscopic residual tumour, R2: macroscopic residual tumour, cT, cN, cM were evaluated by UICC-TNM8th. * Pathological response of Grade 4 was obtained in one patient.

3.1. Surgical outcomes and prognosis

Vascular (HA and PV) resection followed by reconstruction was performed more frequently in NAC-R than in US-R. However, there were no significant differences in terms of operative duration and blood loss. Between US-R and NAC-R, R0 achievement was achieved in 74.2% (23/31) and 69.4% (25/36) of patients, and postoperative complications of C-D grade III or higher were observed in 48.4% (15/31) and 61.1% (22/36), respectively, showing no significant differences. The 90-day mortality rate was zero in both the groups (Table 3). There was no significant difference in OS between US-R and NAC-R (MST 74 months vs. 57 months, p = 0.970), but NAC-UR was significantly poorer than the other two groups (MST 17 months, p < 0.001) (Fig. 2).

Table 3

Comparison of clinical factors in patients with NAC
	R n = 11	BR n = 33	LA n = 15	p
Age, years	77 (62–84)	70 (50–84)	76 (53–84)	0.104
Sex, male/female	7/4	20/13	13/3	0.413
T4/<=T3	1/11	23/10	11/4	< 0.001
N0/1–2	6/5	12/21	7/8	0.530
Bismuth type, I/II/IIIa/IIIb//IV	2/2/2/3/2	0/2/6/11/14	0/0/4/2/9	0.037
Types of NAC
GS/G/GC/GS to GC	9/1/1/0	28/0/2/3	14/1/0/0	0.390
Courses of NAC	2 (1–7)	2 (1–11)	2 (1.5–10)	0.503
Blood examination, initial value
White blood cell, /uL	5680 (3890–7380)	5830 (3060–11810)	5490 (3540–11000)	0.822
Neutrophil, /uL	3400 (1910–5260)	3470 (1420–9220)	3940 (1390–9320)	0.556
Lymphocyte, /uL	1550 (1200–2180)	1580 (600–3040)	1470 (880–2150)	0.565
Platelet count, x10⁴ /uL	19.2 (11.6–35.4)	24.6 (10.2–45.7)	20.6 (12.1–51.1)	0.210
Albumin, g/dL	3.7 (3.0-4.5)	4.0 (2.6–5.1)	3.7 (2.9–4.6)	0.536
NLR	2.17 (1.14–3.60)	2.05 (1.09–7.90)	2.65 (0.77–10.6)	0.333
PNI	45.2 (38.2–52.8)	46.8 (35.2–61.5)	43.1 (35.6–54.8)	0.300
CEA, ng/mL	7.2 (1.9–21.5)	3.5 (0.9–16.3)	3.8 (2.0-11.4)	0.179
CA19-9, U/mL	144 (34.7-10899)	133 (0.2-76452)	469 (5.2–3525)	0.508
US: upfront surgery, NAC: neoadjuvant chemotherapy, TNM8th, R: resectable, BR: borderline resectable, LA: locally advanced, G: gemcitabine, GS: gemcitabine plus S-1, GC: gemcitabine plus cisplatin, NLR: neutrophil-to-lymphocyte ratio, PNI: prognostic nutritional index, CEA: carcinoembryonic antigen, CA19-9: carbohydrate antigen 19 − 9. cT and cN were evaluated by UICC-TNM8th.

3.2. Efficacy and safety of NAC

The resectability of the 59 patients with NAC was R in 11, BR in 33, and LA in 15 patients. When comparing patient characteristics, the rate of cT4 was significantly higher in the LA group, and the rate of Bismuth type I was significantly higher in R (Table 3).

Of the 59 patients who received NAC, 56 (91.7%) completed two courses of NAC (median, 2; range, 1–11) (Table 4). Three patients could not complete NAC due to the following Grade 3 adverse events: skin rashes in two and thrombocytopenia in one. Adverse events of CTCAE Grade 3 or 4 occurred in 10 patients (17.0%): neutropenia in seven (febrile neutropenia in two), anaemia in three, skin rash in two, thrombocytopenia in two, severe diarrhoea in one, pneumonia in one (above, Grade 3), and transient renal failure of Grade 4 managed by haemodialysis in one (overlapped). Cholangitis due to biliary obstruction was identified in 15 patients (25.4%) who were successfully treated with biliary drainage and antibiotics. After NAC, lymphocyte and platelet counts and PNI significantly decreased compared to those before NAC (Table 4). Liver function, represented by KICG and LHL15, was assessed in 54 patients before or after NAC and in 26 patients both before and after NAC (Fig. 3). KICG was not impaired before or after NAC (0.136 (0.072–0.236) and 0.120 (0.049–0.260), p = 0.360). LHL15 was also not impaired before and after NAC: 0.950 (0.870–0.980) and 0.948 (0.871–0.979), respectively (p = 0.840).

Table 4

The change of clinical values before and after NAC in the patients who underwent curative intent surgery
	NAC n = 59
Neoadjuvant chemotherapy
Courses of NAC	2 (1–11)
Completion rate of initial 2 cycles, yes (%)	56 (91.7%)
Toxicity grade 0/1/2/3/4/5	33/5/11/10/0
Patients with adverse events of Grade 3 or higher, n (%)	10 (17.0%)
Cholangitis, yes (%)	15 (25.4%)
	Before NAC	After NAC	p
Blood examination
White blood cell, /uL	5710 (3060–11810)	5420 (2270–11960)	0.071
Neutrophil, /uL	3450 (1390–9320)	3110 (1270–10800)	0.515
Lymphocyte, /uL	1560 (600–3040)	1325 (350–2750)	0.001
Platelet count, x10⁴ /uL	21.2 (10.2–51.1)	18.5 (6.9–38.4)	< 0.001
Albumin, g/dL	3.8 (2.6–5.1)	3.7 (2.4–4.8)	0.085
NLR	2.20 (0.77–10.6)	2.41 (0.625–16.17)	0.071
PNI	46.0 (35.2–61.5)	44.3 (26.7–57.0)	0.013
CEA, ng/mL	3.70 (0.9–21.5)	3.7 (1.0–41.0)	0.703
CA19-9, U/mL	244 (0.2-76452)	91.6 (0.10-14408)	0.132
Liver function*
ICG R15, %. n = 26	12.1% (1.9–31.7%)	12.5% (2.0-39.3%)	0.629
ICG k value. n = 26	0.136 (0.072–0.236)	0.120 (0.049–0.260)	0.360
LHL15. n = 26	0.950 (0.870–0.98000)	0.948 (0.871–0.979)	0.840
NAC: neoadjuvant chemotherapy, NLR: neutrophil-to-lymphocyte ratio, PNI: prognostic nutritional index, CEA: carcinoembryonic antigen, CA19-9: carbohydrate antigen 19 − 9, ICG: indocyanine green retention rate at 15 minutes, LHL15: liver uptake ratio at 15 minutes. * Liver function before and after NAC was precisely evaluated in 26 patients.

The response rate to NAC, in terms of decrease or no change in the tumour size, was higher in R than in BR and LA, although there was no statistical difference: 100% (11/11) in R vs. 66.2% (11/33) in BR and 60.0% (9/15) in LA (p = 0.058). The resection rate was significantly lower in the LA (33.3%, 5/15) than in the R (72.7%, 8/11) and BR (69.7%, 23/33) groups (p = 0.039) (Table 5). In operative procedures, the combined PV resection and reconstruction rate was significantly higher in the BR group (69.6%, 16/33) than in the other two groups (p = 0.007). There were no significant differences in R0 achievement and postoperative complication rates between R, BR, and LA: R0 was 75.0% (6/8), 65.2% (15/23), and 80.4% (4/5) (p = 0.751), and postoperative complications were 75.0% (6/8), 60.9% (14/23), and 40.0% (2/5), respectively (p = 0.452).

Table 5

Clinical courses of patients with NAC
	R (n = 11)	BR (n = 33)	LA (n = 15)		P
Tumour size response to NAC, yes (%)	11 (100%)	22 (66.7%)	9 (60.0%)		0.058
decrease/no change/increase	4/7/0	11/12/10	3/6/6		0.181
Resection	8 (72.7%)	23 (69.7%)	5 (33.3%)		0.039
Operative duration, min	597 (459–714)	600 (383–728)	633 (395–663)		0.862
Blood loss, mL	1951 (719–3948)	1459 (160–6059)	2343 (1443–2911)		0.397
Operative procedures
S1,2,3,4	3	14	2		0.128
S1,2,3,4,5,8	0	2	0
S1,5,6,7,8	3	7	3
S1,4,5,6,7,8	0	0	0
Bile duct resection alone	2	0	0
Combined PD, yes (%)	0	1	0		0.748
Combined PV resection and reconstruction, yes (%)	1 (12.5%)	16 (69.6%)	1 (20.0%)		0.007
Combined HA resection and reconstruction, yes (%)	0	7 (30.4%)	1 (20.0%)		0.202
pT4 (%)	0	7 (30.4%)	1 (20.0%)		0.202
pN1-2 (%)	2 (25%)	12 (52.2%)	3 (60.0%)		0.343
pM1 (%)	0	2 (8.7%)	1 (20.0%)		0.444
R0 (%)	6 (75.0%)	15 (65.2%)	4 (80.0%)		0.751
Complication (Clavien-Dindo ≧ III) (%)	6 (75.0%)	14 (60.9%)	2 (40.0%)		0.452
Hospital stays	42 (19–161)	36 (13–136)	41 (13–115)		0.558
Un-resection	3 (27.3%)	10 (30.3%)	10 (66.7%)		0.039
Causes of un-resection
Tumour progression	1 (9.1%)	5 (15.2%)	4 (26.7%)		0.457
	Tumour marker increasing n = 1	Liver metastasis n = 2 Peritoneal dissemination n = 1 Progressed to LA n = 2		Peritoneal dissemination n = 1 Further local progression n = 3
Insufficient future remnant liver volume	2	3	5
Refusal of surgery	0	1	1
Refractory cholangitis with liver abscess	0	1	0
NAC: neoadjuvant chemotherapy, R: resectable, BR: borderline resectable, LA: locally advanced unresectable, S: segment, PD: pancreaticoduodenectomy, PV: portal vein, HA: hepatic artery, pT, pN, and pM were evaluated by UICC-TNM8th.

The un-resection rates after NAC were significantly higher in the LA group than in the R and BR groups (10/15 (66.7%) vs. 3/11 (27.3%) vs. 10/33 (30.3%), p = 0.039). Tumour progression after NAC was identified in 3.45% (1/29) of R, 15.2% (5/33) of BR, and 26.7% (4/15) of LA patients. In the BR group, three patients developed distant metastasis and two developed local progression after NAC, resulting in un-resection. In the LA group, three patients developed further local advancement and one developed peritoneal dissemination which could not be converted to resection. The future remnant liver volume, estimated using f-rem-KICG, was insufficient, and curative intent surgery was not indicated in 18.2% (2/11) of R, 9.1% (3/33) of BR, and 33.3% (5/15) of LA.

Multivariate analysis revealed that LA (HR 7.54, CI 1.214–46.831, p = 0.030) and age (HR 1.09, CI 1.006–1.183, p = 0.038) were significant risk factors for un-resection after NAC (Table 6) [16].

Table 6

Risk factors of un-resection after NAC
	Univariate analysis			Multivariate analysis
	Resection n = 36	Un-resection n = 23	P	HR	95% CI	P
Age, years	70 (50–84)	76 (61–81)	0.029	1.091	1.006–1.183	0.036
Sex, male/female	22/14	17/6	0.311
Respectability						0.061
R	8	3	0.039	1.000		-
BR	23	10		1.936	0.383–9.801	0.160
LA	5	10		7.539	1.214–46.831	0.030
Bismuth, type1-3/4	14/22	7/16	0.498
cT, T4/T3	19/17	16/7	0.200
cN0/1–2	21/15	13/10	0.891
NAC, GS/G/GC/GS to GC	31/1/2/2	20/1/1/1	0.980
NAC courses	2 (1–11)	2 (1–7)	0.431
CTCAE grade 2 or lower/ 3 or higher	32/4	17/6	0.135
Cholangitis, yes/no	11/25	4/19	0.257
Blood Test, before NAC
White blood cell, /uL	6035 (3060–11810)	5490 (3540–11000)	0.192
Neutrophil, /uL	3555 (1420–9220)	3090 (1390–9320)	0.090
Lymphocyte, /uL	1565 (990–3040)	1480 (600–2440)	0.721
Platelet count, x10⁴ /uL	20.9 (10.2–51.1)	22.3 (11.6–45.7)	0.876
Albumin, g/dL	3.8 (3.0-5.1)	3.9 (2.6-.6)	0.969
NLR	2.54 (1.19–4.78)	2.05 (0.77–10.6)	0.150
PNI	46.2 (38.1–61.5)	44.4 (35.2–56.2)	0.573
CEA, ng/mL	3.8 (0.9–21.5)	3.60 (1.1–11.9)	0.834
CA19-9, U/mL	116 (0.2-76452)	766 (5.2–9066)	0.037
US: upfront surgery, NAC: neoadjuvant chemotherapy, R: resectable, BR: borderline resectable, LA: locally advanced unresectable, G: gemcitabine, GS: gemcitabine plus S-1, GC: gemcitabine plus cisplatin, CTCAE: Common Toxicity Criteria for Adverse Events, NLR: neutrophil-to-lymphocyte ratio, PNI: prognostic nutritional index, CEA: carcinoembryonic antigen, CA19-9: carbohydrate antigen 19 − 9, cT and cN were evaluated by UICC-TNM8th.

In this study, we aimed to increase the resection rate and improve the prognosis of PHC by selecting US and NAC followed by resection based on our resectability criteria, and confirmed the following: 1) curative intent surgery after NAC for advanced PHC presented a similar prognosis compared with US for R-PHC; 2) NAC could be conducted safely without impairing liver function and causing severe adverse events and postoperative complications; 3) compared to R, LA was a significant risk factor for non-resection even after NAC; 4) NAC could prevent tumour progression in most cases, and the efficacy was identified more frequently in R than in BR and LA.

Surgical resection, which is the only potentially curative treatment, is not often indicated because PHC is frequently diagnosed at an advanced stage, with distant metastases and local invasion of the HA and PV [17]. Approximately 60% of PHC cases are diagnosed as unresectable at the initial diagnosis [2, 4, 16]. However, chemotherapy may downsize the tumour and allow conversion surgery for locally advanced PHC. Kato et al. reported that preoperative chemotherapy enabled the downsizing of an initially unresectable locally advanced cholangiocarcinoma and contributed to radical resection in a particular proportion of patients [11]. Tumour downsizing was observed in 22 of the 29 patients with locally advanced diseases, and surgical resection was performed in eight patients (36.7%) with intrahepatic cholangiocarcinoma or gallbladder carcinoma. Yadav et al. demonstrated that neoadjuvant chemotherapy followed by surgery provided better OS in a retrospective study regarding cholangiocarcinoma compared to upfront resection followed by AC (median OS:40.3 vs 32.8 months; HR, 0.78; 95% CI, 0.64–0.94; p = 0.01) [9]. However, most of the diseases in that study were intrahepatic cholangiocarcinomas and gallbladder carcinomas. The efficacy and safety of NAC for the treatment of PHC therefore remains unclear.

We performed NAC for locally advanced PHC based on our resectability criteria [10]. R with regional LN metastases was treated with NAC as an advanced PHC because LN metastases are associated with poor prognosis even after resection [17–19]. BR was also treated by NAC to decrease the risk for positive margin resection, which resulted in early recurrence and a poor prognosis, presuming curative resection [17, 20]. LA was treated with chemotherapy to evaluate the timing of conversion surgery for curative resection. Although US-R was performed for less advanced PHC compared to NAC-R, the R0 achievement rate (74.2% vs. 69.4%) and OS rates were similar (5-year survival: 51.6% vs. 48.4%) and comparable to those of previous reports [1, 20]. NAC-R therefore promises favourable surgical outcomes for advanced PHC. However, the surgical conversion rate after chemotherapy was 61% (36/59) in our patients with advanced PHC. The tumour size response to NAC, which was defined as a decrease or no change in tumour size, was obtained more frequently in R (100%) than in BR (66.7%) and LA (60.0%) patients. These results suggest that NAC can control tumour progression at an earlier stage. Yadav et al. identified that patients with stage I cholangiocarcinoma may be more likely to benefit from NAC than those with stage II-III cholangiocarcinoma [9]. Furthermore, accurate diagnosis of regional LN metastases is difficult. Although 31 patients were diagnosed as clinically lymph node-negative using PET-CT and underwent US, pathological assessment was positive in 14 patients (45%). The survival of LN node-positive PHC patients is poor with surgery alone, and often requires an AC supplement [18, 21]. However, not all patients receive AC depending on their postoperative conditions. Surgical resection requires high-level techniques of major hepatectomy with concomitant vascular resection and reconstruction, which are highly associated with life-threatening postoperative complications [4, 6, 17]. The morbidity and mortality are at about 50% and 2–5% even in high-volume centres [1, 2]. Such postoperative complications disturb the smooth induction of AC. The dose intensity of chemotherapy must be superior to that of AC for NAC. Furthermore, NAC is commonly used in other carcinomas [22–24]. In pancreatic ductal adenocarcinoma in particular, NAC is widely used to downsize the tumour for curative resection in BR and LA status, and to diminish possible micrometastases in R status [25–27]. Accordingly, the routine induction of NAC may contribute to improving OS in PHC, regardless of resectability.

Our resectability criteria were useful for predicting tumour progression and determining surgical indications for advanced PHC after NAC. Tumour progression, such as distant metastases and local progression through NAC, was identified in the order of R, BR, and LA: 9.1%, 15.2%, and 26.7%, respectively. Indeed, local progression from BR to LA was identified in two patients, who might have been successfully treated if US was performed. However, US for BR may result in tumour margin-positive resection, which is highly associated with poor prognosis [17, 20, 28]. Multivariate analysis revealed that BR was not a significant risk factor for un-resection after NAC or R. Therefore, the risk of tumour advancement through NAC in BR was acceptable. Conversely, LA remained a significant risk factor for unresectioning, even after NAC. Our resectability criteria demonstrated not only surgical indications, but also disease activity. NAC might allow for better selection of patients for surgery conversion by excluding patients who postoperatively develop early metastatic recurrence and sparing them from unnecessary surgery. Older age was also revealed as a risk factor for non-resection through NAC. Younger patients are well known to be able to better tolerate aggressive chemotherapy and subsequent surgery.

NAC can be safely conducted without affecting liver function and the postoperative course. As expected, most of the adverse events in this study were haematotoxicity. One patient developed transient renal failure of CTCAE Grade 4 because cholangitis was caused by chemotherapy-induced neutropenia. There are currently no reports in which liver function has been evaluated using KICG and LHL15 through NAC. These are commonly used to assess remnant liver volume and function to prevent postoperative liver failure [12–13]. No patient in our cohort was diagnosed with contraindications for curative resection of impaired liver function (f-rem KICG < 0.5) after NAC. It is a definitive advantage in major surgeries with high complications that NAC does not affect the postoperative course. Patients’ lymphocyte counts, platelet counts, and PNIs were significantly decreased by NAC in our cohort. Although they were associated with systematic inflammation and nutrition, their influence on the clinical course and prognosis was modest because their decrease did not meet the CTCAE 5.0. The safety revealed in this study will hopefully help to reduce the NAC threshold for PHC.

This study has the limitation of being a one-arm retrospective study with a small sample size. Chemotherapy regimens were not fixed, although most were two courses of GS therapy. Therefore, more aggressive chemotherapy regimens are required. Since January 2021, we have conducted three-drug combination chemotherapy using gemcitabine, cisplatin, and S-1 (GCS therapy) to increase the cytotoxic effect (Neoadjuvant GCS therapy for locally advanced perihilar cholangiocarcinoma, jRCTs041210063). GCS therapy has also been conducted without any specific significant side effects, and is expected to have favourable efficacy in advanced PHC. In the future, the use of tyrosine kinase inhibitors based on tumour sequencing data and vascular endothelial growth factor receptor inhibition for regulating angiogenesis may also provide a benefit as a preoperative treatment [5, 29–30]. A multicentre prospective clinical trial is necessary to further explore the role of NAC for PHC.

NAC can be safely performed without impairing liver function, based on our resectability criteria, and contributed to allowing for curative resection in cases of advanced PHC. R-PHC was more responsive to gemcitabine-based NAC than was LA, which remained a strong risk factor for unresectable tumours requiring radical interventions for curative resection.

AC, adjuvant chemotherapy; BR, borderline resectable; NAC, neoadjuvant chemotherapy; NAC-R, NAC followed by resection; NAC-UR, un-resection after NAC; PHC, perihilar cholangiocarcinoma; R, resectable; US, upfront surgery; US-R, resection by upfront surgery; LA, locally advanced unresectable; LN, lymph node.

Funding: The present study was not funded by any organization.

Disclosure of potential conflicts of interest: The authors declare no conflicts of interest.

Ethics approval: The protocol for this research project has been approved by a suitably constituted Ethics Committee of the institution, which conforms to the provisions of the Declaration of Helsinki. Committee of Mie University Hospital, Approval No. 2954.

Consent to participate: Informed consent was obtained via an opt-out on the website.

Contribution: All authors contributed to the conception and design of the study and the interpretation of the data. Gyoten K contributed to the data acquisition and analysis and drafted the manuscript. Gyoten K, Kuriyama N, and Mizuno S critically revised the manuscript for important intellectual content. All authors provided final approval for the version to be published.

Mueller M, Breuer E, Mizuno T, Bartsch F, Ratti F, Benzing C, et al. Perihilar Cholangiocarcinoma - Novel Benchmark Values for Surgical and Oncological Outcomes From 24 Expert Centers. Ann Surg. 2021;274(5):780–8.
Mizuno T, Ebata T, Yokoyama Y, Igami T, Yamaguchi J, Onoe S, et al. Combined Vascular Resection for Locally Advanced Perihilar Cholangiocarcinoma. Ann Surg. 2022;275(2):382–90.
Nagino M. Perihilar cholangiocarcinoma: a surgeon's viewpoint on current topics. J Gastroenterol. 2012;47(11):1165–76.
Di Benedetto F, Magistri P, Di Sandro S. Perihilar-cholangiocarcinoma: what really matters? Hepatobiliary Surg Nutr. 2022;11(4):616–9.
Valle JW, Lamarca A, Goyal L, Barriuso J, Zhu AX. New Horizons for Precision Medicine in Biliary Tract Cancers. Cancer Discov. 2017;7(9):943–62.
Watanabe N, Mizuno T, Yamaguchi J, Yokoyama Y, Igami T, Onoe S, et al. A proposal of drain removal criteria in hepatobiliary resection. J Hepatobiliary Pancreat Sci. 2022;29(9):974–82.
Cillo U, Fondevila C, Donadon M, Gringeri E, Mocchegiani F, Schlitt HJ, et al. Surgery for cholangiocarcinoma. Liver International. 2019;39(S1):143–55.
Kuriyama N, Komatsubara H, Nakagawa Y, Maeda K, Shinkai T, Noguchi D, et al. Impact of Combined Vascular Resection and Reconstruction in Patients with Advanced Perihilar Cholangiocarcinoma. J Gastrointest Surg. 2021;25(12):3108–18.
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.
Kuriyama N, Usui M, Gyoten K, Hayasaki A, Fujii T, Iizawa Y, et al. Neoadjuvant chemotherapy followed by curative-intent surgery for perihilar cholangiocarcinoma based on its anatomical resectability classification and lymph node status. BMC Cancer. 2020;20(1).
Kato A, Shimizu H, Ohtsuka M, Yoshidome H, Yoshitomi H, Furukawa K, et al. Surgical resection after downsizing chemotherapy for initially unresectable locally advanced biliary tract cancer: a retrospective single-center study. Ann Surg Oncol. 2013;20(1):318–24.
Yokoyama Y, Nishio H, Ebata T, Igami T, Sugawara G, Nagino M. Value of indocyanine green clearance of the future liver remnant in predicting outcome after resection for biliary cancer. Br J Surg. 2010;97(8):1260–8.
Sumiyoshi T, Shima Y, Okabayashi T, Kozuki A, Hata Y, Noda Y, et al. Liver function assessment using 99mTc-GSA single-photon emission computed tomography (SPECT)/CT fusion imaging in hilar bile duct cancer: A retrospective study. Surgery. 2016;160(1):118–26.
Iwasaki M, Takada Y, Hayashi M, Minamiguchi S, Haga H, Maetani Y, et al. Noninvasive evaluation of graft steatosis in living donor liver transplantation. Transplantation. 2004;78(10):1501–5.
Miyazaki M, Kimura F, Shimizu H, Yoshidome H, Otsuka M, Kato A, et al. Extensive hilar bile duct resection using a transhepatic approach for patients with hepatic hilar bile duct diseases. Am J Surg. 2008;196(1):125–9.
Rizvi S, Khan SA, Hallemeier CL, Kelley RK, Gores GJ. Cholangiocarcinoma - evolving concepts and therapeutic strategies. Nat Rev Clin Oncol. 2018;15(2):95–111.
DeOliveira ML, Cunningham SC, Cameron JL, Kamangar F, Winter JM, Lillemoe KD, et al. Cholangiocarcinoma: thirty-one-year experience with 564 patients at a single institution. Ann Surg. 2007;245(5):755–62.
Mizuno T, Ebata T, Yokoyama Y, Igami T, Sugawara G, Yamaguchi J, et al. Adjuvant gemcitabine monotherapy for resectable perihilar cholangiocarcinoma with lymph node involvement: a propensity score matching analysis. Surg Today. 2017;47(2):182–92.
Kitagawa Y, Nagino M, Kamiya J, Uesaka K, Sano T, Yamamoto H, et al. Lymph node metastasis from hilar cholangiocarcinoma: audit of 110 patients who underwent regional and paraaortic node dissection. Ann Surg. 2001;233(3):385–92.
Komaya K, Ebata T, Yokoyama Y, Igami T, Sugawara G, Mizuno T, et al. Recurrence after curative-intent resection of perihilar cholangiocarcinoma: analysis of a large cohort with a close postoperative follow-up approach. Surgery. 2018;163(4):732–8.
Takahashi D, Mizuno T, Yokoyama Y, Igami T, Yamaguchi J, Onoe S, et al. Adjuvant S-1 vs gemcitabine for node‐positive perihilar cholangiocarcinoma: A propensity score‐adjusted analysis. J Hepatobiliary Pancreat Sci. 2021;28(9):716–26.
Jung F, Lee M, Doshi S, Zhao G, Cheung KLT, Chesney T, et al. Neoadjuvant therapy versus direct to surgery for T4 colon cancer: meta-analysis. Br J Surg. 2021;109(1):30–6.
van Dam JL, Janssen QP, Besselink MG, Homs MYV, van Santvoort HC, van Tienhoven G, et al. Neoadjuvant therapy or upfront surgery for resectable and borderline resectable pancreatic cancer: A meta-analysis of randomised controlled trials. Eur J Cancer. 2022;160:140–9.
Pasquali S, Yim G, Vohra RS, Mocellin S, Nyanhongo D, Marriott P, et al. Survival After Neoadjuvant and Adjuvant Treatments Compared to Surgery Alone for Resectable Esophageal Carcinoma: A Network Meta-analysis. Ann Surg. 2017;265(3):481–91.
Motoi F, Kosuge T, Ueno H, Yamaue H, Satoi S, Sho M, et al. Randomized phase II/III trial of neoadjuvant chemotherapy with gemcitabine and S-1 versus upfront surgery for resectable pancreatic cancer (Prep-02/JSAP05). Jpn J Clin Oncol. 2019;49(2):190–4.
Motoi F, Unno M. Adjuvant and neoadjuvant treatment for pancreatic adenocarcinoma. Jpn J Clin Oncol. 2020;50(5):483–9.
Birrer DL, Golcher H, Casadei R, Haile SR, Fritsch R, Hussung S, et al. Neoadjuvant Therapy for Resectable Pancreatic Cancer: A New Standard of Care. Pooled Data From 3 Randomized Controlled Trials. Ann Surg. 2021;274(5):713–20.
Lauterio A, De Carlis R, Centonze L, Buscemi V, Incarbone N, Vella I, et al. Current Surgical Management of Peri-Hilar and Intra-Hepatic Cholangiocarcinoma. Cancers (Basel). 2021;13(15):3657.
Churi CR, Shroff R, Wang Y, Rashid A, Kang HC, Weatherly J, et al. Mutation profiling in cholangiocarcinoma: prognostic and therapeutic implications. PLoS One. 2014;9(12):e115383.
Lubner SJ, Mahoney MR, Kolesar JL, Loconte NK, Kim GP, Pitot HC, et al. Report of a multicenter phase II trial testing a combination of biweekly bevacizumab and daily erlotinib in patients with unresectable biliary cancer: a phase II Consortium study. J Clin Oncol. 2010;28(21):3491–7.

No competing interests reported.

Download PDF

Journal Publication

published 01 Jul, 2023

Read the published version in Langenbeck's Archives of Surgery →

Editorial decision: Major revision
15 Apr, 2023
Reviews received at journal
24 Mar, 2023
Reviewers agreed at journal
18 Mar, 2023
Reviewers invited by journal
26 Feb, 2023
Editor assigned by journal
18 Jan, 2023
Submission checks completed at journal
18 Jan, 2023
First submitted to journal
17 Jan, 2023

You are reading this latest preprint version

The safety and efficacy of neoadjuvant chemotherapy based on our resectability criteria for locally advanced perihilar cholangiocarcinoma

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Methods

2.1. Patients

2.2. Our resectability criteria

2.3. NAC

2.4. Indication of curative intent surgery

2.5. Evaluation of perioperative clinical courses

2.6. Adjuvant chemotherapy

2.7. Statistical analysis

3. Results

3.1. Surgical outcomes and prognosis

3.2. Efficacy and safety of NAC

4. Discussion

Abbreviations

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1