Imaging of the hepatic arterial infusion pump: Primer for radiologists.

Hepatic arterial infusion (HAI) pumps are used to deliver liver-directed therapy by allowing the administration of selective chemotherapy to the liver via a catheter implanted most commonly into the gastroduodenal artery connected to a subcutaneous pump. This selective administration helps maximize the chemotherapeutic effect within the hepatic tumors while minimizing systemic toxicity. While HAI therapy has primarily been used to treat liver-only metastatic colorectal cancer, the indications have expanded to other malignancies, including intrahepatic cholangiocarcinoma. Radiologists play an important role in pre-operative planning, assessment of treatment response, and evaluation for potential complications using various imaging studies, including computed tomography angiography, magnetic resonance imaging, and perfusion scintigraphy. This article describes the radiologist's role as part of a multi-disciplinary oncology team to help maximize the success of HAI therapy and also helps radiologists familiarize themselves with various aspects of HAI pumps.

Hepatectomy is associated with improved oncologic outcomes in recurrent colorectal liver metastases: A propensity-matched analysis.

BACKGROUND: Following resection of colorectal liver metastasis, most patients have disease recurrence, most commonly intrahepatic. Although the role of resection in colorectal liver metastasis is well-established, there have been limited investigations assessing the benefit of repeat hepatic resection compared with systemic treatment alone for intrahepatic recurrence. METHODS: A retrospective single-institution cohort study of patients with recurrent colorectal liver metastasis following curative-intent hepatectomy was performed from 2003 to 2019. The oncologic outcomes, including post-recurrence overall survival, were evaluated using Kaplan-Meier and Cox proportional hazards modeling. Patients undergoing repeat hepatic resection were propensity-matched with patients receiving systemic treatment alone based on relevant clinicopathologic variables. RESULTS: There were 338 patients treated with hepatic resection for colorectal liver metastasis over the study period. Liver recurrence was observed in 147 (43%) patients at a median time of 10 months from prior resection, with a median post-recurrence overall survival of 29 months. There were 37 patients managed with repeat hepatic resection; 33 (89%) received perioperative chemotherapy. On propensity matching, there were no significant clinicopathologic differences between 37 patients having repeat hepatic resection and 37 patients treated with systemic treatment alone. Repeat hepatic resection was independently associated with improved 5-year post-recurrence overall survival compared with systemic treatment alone (median overall survival 41 vs 35 months, 5-year overall survival 19% vs 3%, P = .048). CONCLUSION: Disease characteristics of patients with intrahepatic recurrence of colorectal liver metastasis, specifically the number of liver lesions and size of the largest lesion, are most predictive of survival and response to systemic therapy. Patients who recur with oligometastatic liver disease experience improved outcomes and derive benefit from curative-intent repeat hepatic resection with integrated perioperative systemic therapy.

Ten-Year Survivorship in Patients with Metastatic Gastrointestinal Stromal Tumors.

INTRODUCTION: Patients developing metastatic gastrointestinal stromal tumors (mGIST) have heterogenous disease biology and oncologic outcomes; prognostic factors are incompletely characterized. We sought to evaluate predictors of 10-year metastatic survivorship in the era of tyrosine kinase inhibitor (TKI) therapy. METHODS: We reviewed patients with mGIST treated at our Comprehensive Cancer Center from 2003 to 2019, including only patients with either mortality or 10 years of follow-up. Ten-year survivorship was evaluated with logistic regression. RESULTS: We identified 109 patients with a median age of 57 years at mGIST diagnosis. Synchronous disease was present in 57% (n = 62) of patients; liver (n = 48, 44%), peritoneum (n = 40, 37%), and liver + peritoneum (n = 18, 17%) were the most common sites. Forty-six (42%) patients were 10-year mGIST survivors. Following mGIST diagnosis, radiographic progression occurred within 2 years in 53% (n = 58) of patients, 2-5 years in 16% (n = 17), and 5-10 years in 16% (n = 17), with median survival of 32, 76, and 173 months, respectively. Seventeen (16%) patients had not progressed by 10 years. Fifty-two (47%) patients underwent metastasectomy, which was associated with improved progression-free survival (hazard ratio 0.63, p = 0.04). In patients experiencing progression, factors independently associated with 10-year survivorship were age (odds ratio [OR] 0.96, p = 0.03) and time to progression (OR 1.71/year, p < 0.001). CONCLUSIONS: Ten-year survivorship is achievable in mGIST in the era of TKIs and is associated with younger age and longer time to first progression, while metastasectomy is associated with longer time to first progression. The role of metastasectomy in the management of patients with disease progression receiving TKI therapy merits further study.

Hepatic arterial infusion pump chemotherapy combined with systemic therapy for patients with advanced colorectal liver metastases: Outcomes in a newly established program.

BACKGROUND AND OBJECTIVES: Colorectal liver metastasis (CRLM) is a leading cause of morbidity and mortality in patients with colorectal cancer. Hepatic arterial infusion (HAI) chemotherapy has been demonstrated to improve survival in patients with resected CRLM and to facilitate conversion of technically unresectable disease. METHODS: Between 2016 and 2018, n = 22 HAI pumps were placed for CRLM. All patients received systemic chemotherapy concurrently with HAI floxuridine/dexamethasone. Overall survival (OS) and progression-free survival (PFS) were assessed using the Kaplan-Meier method. RESULTS: HAI pumps were placed in seven patients with completely resected CRLM and 15 patients with unresectable disease. Twenty-one patients received HAI floxuridine with a median of 5 total HAI cycles (interquartile range: 4-7). Biliary sclerosis was the most common HAI-related complication (n = 5, 24%). Of the 13 patients treated to convert unresectable CRLM, 3 (23%) underwent hepatic resection with curative intent after a median of 7 HAI cycles (range: 4-10). For all HAI patients, the mean OS was 26.7 months from CRLM diagnosis, while the median PFS and hepatic PFS from pump placement were 9 and 13 months, respectively. CONCLUSION: Concomitant HAI and systemic therapy can be utilized at multidisciplinary programs for patients with advanced CRLM, both in the adjuvant setting and to facilitate conversion of unresectable disease.

Surgical timing after preoperative chemotherapy is associated with oncologic outcomes in resectable colorectal liver metastases.

INTRODUCTION: Preoperative chemotherapy (POC) is often employed for patients with resectable colorectal liver metastasis (CRLM). The time to resection (TTR) following the end of chemotherapy may impact oncologic outcomes; this phenomenon has not been studied in CRLM. METHODS: We queried our institutional cancer database for patients with resected CRLM after POC from 2003 to 2019. TTR was calculated from date of last cytotoxic chemotherapy. Kaplan-Meier analysis and multivariable Cox proportional hazards modeling were used to analyze recurrence-free survival (RFS) and overall survival (OS). RESULTS: We identified n = 187 patients. One hundred twenty-four (66%) patients had a TTR of <2 months, while 63 (33%) had a TTR of ≥2 months. Median follow-up was 36 months. On Kaplan-Meier analysis, patients with TTR ≥ 2 months had shorter RFS (median 11 vs. 17 months, p = 0.002) and OS (median 44 vs. 62 months, p < 0.001). On multivariable analysis, TTR ≥ 2 months was independently associated with worse RFS (hazard ratio [HR] = 1.54, 95% confidence interval [CI] = 1.06-2.22, p = 0.02) and OS (HR = 1.75, 95% CI = 1.11-2.77, p = 0.01). CONCLUSION: TTR ≥ 2 months following POC is independently associated with worse oncologic outcomes in patients with resectable CRLM. We therefore recommend consideration for hepatic resection of CRLM within this window whenever feasible.

Hepatic metastases in gastrointestinal stromal tumors: oncologic outcomes with curative-intent hepatectomy, resection of treatment-resistant disease, and tyrosine kinase inhibitor therapy alone.

BACKGROUND: Hepatic resection for metastatic GIST (mGIST) is often performed with either curative-intent or for tyrosine kinase inhibitor (TKI)-resistant lesions. The efficacy of hepatectomy for treatment-resistant lesions (TRL) is uncertain. METHODS: We reviewed patients with liver-mGIST treated from 2003 to 2018. Oncologic outcomes including overall (OS), post-operative progression-free survival (PFS), and post-progression OS were evaluated using Kaplan-Meier and Cox proportional hazards modeling. RESULTS: We identified n = 91 patients; 31 (34%) underwent curative-intent hepatectomy, 60 (66%) were initially managed with TKI alone, and 17 (19%) had resection of a TRL. The median follow-up for resected patients was 102 months (range 5-209 months) with 23 (25%) managed with a major hepatectomy. Patients having curative-intent hepatectomy had 72% 10-year OS following diagnosis of liver-mGIST, compared with 58% (P = 0.50) for TRL resection and 41% (P = 0.01) for non-resected patients. Curative-intent hepatectomy (HR 0.39, P = 0.03) and age (HR 1.04, P = 0.004) were independently associated with 10-year OS, but not TRL resection. TRL resection was not associated with improved post-progression OS compared to second-line TKI therapy (HR 0.61, P = 0.21). CONCLUSIONS: Curative-intent hepatectomy is associated with improved OS in liver-mGIST. The oncologic benefit of resecting treatment-resistant liver-mGIST compared to second-line TKI therapy alone remains unclear in the era of multi-line TKI therapy.

Label-free enrichment of rare unconventional circulating neoplastic cells using a microfluidic dielectrophoretic sorting device.

Cellular circulating biomarkers from the primary tumor such as circulating tumor cells (CTCs) and circulating hybrid cells (CHCs) have been described to harbor tumor-like phenotype and genotype. CHCs are present in higher numbers than CTCs supporting their translational potential. Methods for isolation of CHCs do not exist and are restricted to low-throughput, time consuming, and biased methodologies. We report the development of a label-free dielectrophoretic microfluidic platform facilitating enrichment of CHCs in a high-throughput and rapid fashion by depleting healthy peripheral blood mononuclear cells (PBMCs). We demonstrated up to 96.5% depletion of PBMCs resulting in 18.6-fold enrichment of cancer cells. In PBMCs from pancreatic adenocarcinoma patients, the platform enriched neoplastic cells identified by their KRAS mutant status using droplet digital PCR with one hour of processing. Enrichment was achieved in 75% of the clinical samples analyzed, establishing this approach as a promising way to non-invasively analyze tumor cells from patients.

Geographic Disparities in Referral Rates and Oncologic Outcomes of Intrahepatic Cholangiocarcinoma: A Population-Based Study.

BACKGROUND: Intrahepatic cholangiocarcinoma (ICC) is a rare cancer. Patients in rural areas may face reduced access to advanced treatments often only available at referral centers. We evaluated the association of referral center treatment with treatment patterns, outcomes, and geography in patients with ICC. METHODS: We queried the Oregon State Cancer Registry for ICC between 1997 and 2016, collecting clinicopathologic, demographic, and oncologic data. Patients were classified by treatment at a referral center or non-referral center. 'Crowfly' distance to the nearest referral center (DRC) was calculated. Outcomes were evaluated using Kaplan-Meier, Cox proportional hazards modeling, and logistic regression. RESULTS: Over 20 years, 740 patients with ICC had a median age of 66 years. Slightly more than half (n = 424, 57%) were non-referral center treated and 316 (43%) were referral center treated. Referral center treatment increased over time (odds ratio [OR] 1.03/year, p < 0.05). Referral center-treated patients had improved overall survival in all patients (median 9 vs. 4 months, p < 0.001), in the non-metastatic group (median 13 vs. 6 months, p < 0.001), and in patients not receiving liver resection (median 6 vs. 3 months, p < 0.05). On multivariable analysis, referral center-treated patients more often underwent chemotherapy, resection, or radiation (all p < 0.05). Increasing DRC (OR 0.98/20 km, p < 0.05) was independently associated with non-referral center treatment. CONCLUSION: Patients with ICC who are evaluated at a referral center are more likely to receive treatments associated with better oncologic outcomes, including patients who are not managed with hepatic resection. Increasing the DRC is associated with treatment at a non-referral center; interventions to facilitate referral, such as telemedicine, may lead to improved outcomes for patients with ICC in rural states.

Circulating Hybrid Cells: A Novel Liquid Biomarker of Treatment Response in Gastrointestinal Cancers.

BACKGROUND: Real-time monitoring of treatment response with a liquid biomarker has potential to inform treatment decisions for patients with rectal adenocarcinoma (RAC), esophageal adenocarcinoma (EAC), and colorectal liver metastasis (CRLM). Circulating hybrid cells (CHCs), which have both immune and tumor cell phenotypes, are detectable in the peripheral blood of patients with gastrointestinal cancers, but their potential as an indicator of treatment response is unexplored. METHODS: Peripheral blood specimens were collected from RAC and EAC patients after neoadjuvant therapy (NAT) or longitudinally during therapy and evaluated for CHC levels by immunostaining. Receiver operating characteristics (ROCs) and the Kaplan-Meier method were used to analyze the CHC level as a predictor of pathologic response to NAT and disease-specific survival (DSS), respectively. RESULTS: Patients with RAC (n = 23) and EAC (n = 34) were sampled on the day of resection, and 11 patients (32%) demonstrated a pathologic complete response (pCR) to NAT. On ROC analysis, CHC levels successfully discriminated pCR from non-pCR with an area under the curve of 0.82 (95% confidence interval [CI], 0.71-0.92; P < 0.001). Additionally, CHC levels in the EAC patients correlated with residual nodal involvement (P = 0.026) and 1-year DSS (P = 0.029). The patients with RAC who were followed longitudinally during NAT (n = 2) and hepatic arterial infusion therapy for CRLM (n = 2) had CHC levels that decreased with therapy response and increased before clinical evidence of disease progression. CONCLUSION: Circulating hybrid cells are a novel blood-based biomarker with potential for monitoring treatment response and disease progression to help guide decisions for further systemic therapy, definitive resection, and post-therapy surveillance. Additional validation studies of CHCs are warranted.

Relevance of circulating hybrid cells as a non-invasive biomarker for myriad solid tumors.

Metastatic progression defines the final stages of tumor evolution and underlies the majority of cancer-related deaths. The heterogeneity in disseminated tumor cell populations capable of seeding and growing in distant organ sites contributes to the development of treatment resistant disease. We recently reported the identification of a novel tumor-derived cell population, circulating hybrid cells (CHCs), harboring attributes from both macrophages and neoplastic cells, including functional characteristics important to metastatic spread. These disseminated hybrids outnumber conventionally defined circulating tumor cells (CTCs) in cancer patients. It is unknown if CHCs represent a generalized cancer mechanism for cell dissemination, or if this population is relevant to the metastatic cascade. Herein, we detect CHCs in the peripheral blood of patients with cancer in myriad disease sites encompassing epithelial and non-epithelial malignancies. Further, we demonstrate that in vivo-derived hybrid cells harbor tumor-initiating capacity in murine cancer models and that CHCs from human breast cancer patients express stem cell antigens, features consistent with the potential to seed and grow at metastatic sites. Finally, we reveal heterogeneity of CHC phenotypes reflect key tumor features, including oncogenic mutations and functional protein expression. Importantly, this novel population of disseminated neoplastic cells opens a new area in cancer biology and renewed opportunity for battling metastatic disease.

Degree of biliary tract violation during treatment of gallbladder adenocarcinoma is independently associated with development of peritoneal carcinomatosis.

BACKGROUND: Gallbladder cancer (GBC) is often incidentally diagnosed after cholecystectomy. Intra-operative biliary tract violations (BTV) have been recently associated with development of peritoneal disease (PD). The degree of BTV may be associated with PD risk, but has not been previously investigated. METHODS: We reviewed patients with initially non-metastatic GBC treated at our institution from 2003 to 2018. Patients were grouped based on degree of BTV during their treatment: major (e.g., cholecystotomy with bile spillage, n = 27, 29%), minor (e.g., intra-operative cholangiogram, n = 18, 19%), and no violations (n = 48, 55%). Overall survival (OS) and peritoneal disease-free survival (PDFS) were evaluated with Kaplan-Meier and Cox proportional hazards modeling. RESULTS: Ninety-three patients were identified; the median age was 64 years (range 31-87 years). Seventy-six (82%) were incidentally diagnosed. The median follow-up was 23 months; 20 (22%) patients developed PD. The 3-year PDFS for patients with major, minor, and no BTV was 52%, 83%, and 98%, respectively (major vs. none: p < 0.001; minor vs. none: p < 0.01). BTV was not associated with 5-year OS (HR 1.53, p = 0.16). CONCLUSION: Increasing degree of BTV is associated with higher risk of peritoneal carcinomatosis in patients with GBC and should be considered during preoperative risk stratification. Reporting biliary tract violations during cholecystectomy is encouraged.

Imatinib-resistant gastrointestinal stromal tumors in the era of second- and third-line tyrosine kinase inhibitors: Does surgical resection have a role?

BACKGROUND: Imatinib resistance is associated with a poor prognosis in patients with gastrointestinal stromal tumors. Although novel tyrosine kinase inhibitors have improved outcomes in imatinib-resistant gastrointestinal stromal tumors, the role of resection remains unclear. We sought to investigate factors predictive of overall and progression-free survival in patients with imatinib-resistant gastrointestinal stromal tumors. METHODS: A query of our prospectively maintained Comprehensive Cancer Center registry was performed from 2003 to 2019 for patients with imatinib-resistant gastrointestinal stromal tumors. Clinicopathologic characteristics and medical and surgical treatments were collected; overall survival and progression-free survival after imatinib-resistance were analyzed with Kaplan-Meier and Cox proportional hazards modeling. RESULTS: A total of 84 patients developed imatinib resistance at a median age of 59 years. Median time to imatinib resistance after diagnosis and overall survival after imatinib resistance was 50 and 51 months, respectively. After being diagnosed with imatinib resistance, 17 (20%) patients underwent resection. On multivariable analysis, resection after imatinib resistance was independently associated with improved progression-free survival (hazard ratio 0.50; P = .027) but not overall survival (hazard ratio 0.62; P = .215). Similar findings were found on subgroup analysis of patients treated with second-line sunitinib (n = 71). CONCLUSION: Long-term survival can be achieved in patients who develop imatinib-resistant gastrointestinal stromal tumors. Surgical resection of imatinib-resistant gastrointestinal stromal tumors is associated with improved progression-free survival and should be considered in selected patients.

Disease-free Interval Is Associated with Oncologic Outcomes in Patients with Recurrent Gastrointestinal Stromal Tumor.

BACKGROUND: Gastrointestinal stromal tumors (GIST) commonly recur following curative-intent resection. Patients with recurrent GIST display heterogeneous outcomes with limited prognostic tools. We investigated factors associated with post-recurrence survival (PRS) and progression-free survival (PFS). METHODS: We performed a review of our institutional cancer registry from 2003 to 2018 for patients with GIST. Clinicopathologic and outcome data were collected. The disease-free interval (DFI) was calculated from the end of curative-intent oncologic therapy until recurrence. Outcomes were evaluated using Kaplan-Meier and Cox proportional hazards modeling. RESULTS: Overall, 254 patients underwent resection of primary, non-metastatic GIST, with 81 (32%) recurrences. The median age was 58 years and more than half of the patients with recurrence (n = 44; 54%) received adjuvant imatinib. Recurrence was most common in the liver (n = 34, 42%), peritoneum (n = 31, 38%), or liver plus peritoneum (n = 10, 12%). The median DFI was 14 months (interquartile range 2-26 months); 51 (63%) patients had a DFI ≤24 months and 30 (37%) had a DFI > 24 months. The median post-recurrence follow-up was 46 months. Compared with a DFI ≤24 months, patients with a DFI >24 months had increased 10-year PRS (77% vs. 41%, p < 0.05) and 10-year PFS (73% vs. 19%, p < 0.001). On multivariable analysis controlling for mutational and clinicopathologic features, a DFI >24 months was independently associated with increased PRS (hazard ratio [HR] 0.24, p < 0.01) and PFS (HR 0.18, p < 0.001). CONCLUSIONS: The DFI is independently associated with oncologic outcomes in recurrent GIST and may be useful in treatment planning. Recurrence after 24 months may signify indolent disease biology that may benefit from additional treatment, including metastasectomy.

Circulating hybrid cells predict presence of occult nodal metastases in oral cavity carcinoma.

BACKGROUND: Levels of circulating hybrid cells (CHCs), a newly identified circulating tumor cell (CTC), correlate with disease stage and progression in cancer. We investigated their utility to risk-stratify patients with clinically N0 (cN0) oral cavity squamous cell carcinoma (OCSCC), and to identify patients with occult cervical lymph node metastases (pN+). METHODS: We analyzed peripheral blood samples for CHCs with co-expression of cytokeratin (tumor) and CD45 (leukocyte) from 22 patients with cN0 OCSCC using immunofluorescence microscopy, then correlated levels with pathologic lymph node status. RESULTS: CHC levels exceeded CTCs and correlated with the presence of both clinically overt (p = 0.002) and occult nodal metastases (p = 0.006). CONCLUSIONS: For evaluated cN0 OCSCC patients, those with cN0 → pN+ status harbored elevated CHC levels compared to patients without occult disease. Our findings highlight a promising blood-based biologic assay with potential utility to determine the necessity of surgical neck dissection for staging and treatment.

Neoadjuvant chemotherapy is associated with improved survival in patients undergoing hepatic resection for intrahepatic cholangiocarcinoma.

BACKGROUND: The impact of neoadjuvant chemotherapy (NAC) on overall and recurrence-free survival (OS, RFS) in resectable intrahepatic cholangiocarcinoma (ICC) is poorly characterized. We sought to investigate the association of NAC with oncologic outcomes in ICC. METHODS: We identified n = 52 patients with ICC undergoing hepatectomy from 2004 to 2017. Oncologic outcomes were analyzed using Kaplan-Meier and multivariate Cox proportional hazard modeling. RESULTS: The median patient age was 64-years. NAC was administered in ten (19%) patients, most commonly with gemcitabine-cisplatin (n = 8, 80%). Median RFS and OS were 15 months. and 49 months, respectively. Controlling for stage and margins, NAC was independently associated with improved OS (HR 0.16, P = 0.01) but not RFS (HR 0.54, P = 0.27). NAC was not associated with major post-operative complications (P = 0.25) or R1 margins (P = 0.58). CONCLUSION: NAC in ICC may hold oncologic benefits beyond downstaging borderline resectable disease, such as identifying patients with favorable biology who are more likely to benefit from resection.

Conventional hepatic arterial anatomy? Novel findings and insights of a multi-disciplinary hepatic arterial infusion pump program.

INTRODUCTION: Variant hepatic arterial anatomy (vHAA) is thought to occur in 20-30% of patients. Hepatic arterial infusion (HAI) pump placement for liver cancers requires thorough hepatic artery dissection; we sought to compare vHAA identified during pump placement with established dogma. METHODS: Between 2016 and 2020, n = 30 patients received a HAI pump. Intra-operatively identified vHAA was characterized and compared with published data. RESULTS: vHAA was identified in 60% (n = 18) of patients, significantly higher than 19% (3671 of 19013) in the largest published series (P < 0.001). The most common variations were accessory left (n = 12; 40%) and replaced right (n = 6; 20%) hepatic arteries; six (20%) had ≥2 variants. Pre-operative imaging correctly identified 67% of variant hepatic arteries. DISCUSSION: Meticulous operative dissection of the hepatic arterial tree reveals vHAA not captured by imaging or cadaveric dissection. vHAA likely has a higher prevalence than previously reported and should be addressed to optimize therapeutic efficacy of HAI pump therapy.

Surgical and oncologic outcomes following repeat hepatic resection of colorectal liver metastasis: Who benefits?

BACKGROUND: Resected colorectal liver metastases (CRLM) frequently recur intrahepatically. Selection criteria for repeat hepatectomy of recurrent CRLM are ill-defined. METHODS: We performed an institutional review of patients with recurrent CRLM undergoing repeat hepatectomy from 2003 to 19. Post-recurrence overall (rOS) and recurrence-free survival (RFS) were analyzed with Cox proportional hazards modeling. RESULTS: n = 147 experienced recurrent CRLM; 11% (n = 38) received repeat hepatectomy of which there was one Clavien-Dindo IIIa complication. Median rOS was 41 months; median RFS was 9 months. Improved rOS and RFS were independently associated with additional post-operative chemotherapy after repeat hepatectomy (HR 0.35 and 0.34, respectively); poor rOS with recurrent CRLM >3 cm (HR 4.4) and <12 months from first hepatectomy to recurrence (HR 4.8); poor RFS with ≥3 recurrence liver metastases (HR 2.8) (All P < 0.05). DISCUSSION: Repeat hepatectomy for recurrent CRLM can be performed safely. Worse survival following repeat hepatectomy is independently associated with >3 cm and ≥3 liver lesions at recurrence, and <12 months to recurrence. Additional post-operative chemotherapy after repeat hepatectomy is associated with improved outcomes.

Detection of Tumor Multifocality in Resectable Intrahepatic Cholangiocarcinoma: Defining the Optimal Pre-operative Imaging Modality.

BACKGROUND: Multiple tumor foci (MTF) in intrahepatic cholangiocarcinoma (ICC), including satellitosis and true multifocality, is a known negative prognostic factor and can inform pre-operative decision-making. Lack of standardized pre-operative liver staging practices may contribute to undiagnosed MTF and poor outcomes. We sought to investigate the sensitivity of different cross-sectional imaging modalities for MTF at our institution. METHODS: We identified n = 52 patients with ICC who underwent curative-intent resection from 2004 to 2017 in a multidisciplinary hepato-pancreato-biliary cancer program. Timing and modality of pre-operative imaging were recorded. Blinded review of imaging was performed and modalities were evaluated for false-negative rate (FNR) in detecting MTF, satellitosis, and true multifocality. RESULTS: Forty-one (79%) patients underwent CT and 20 (38%) underwent MRI prior to hepatectomy. MTF was pre-operatively identified in six (12%) patients. An additional seven patients had MTF discovered on final surgical pathology, despite a median interval from CT/MRI to surgery of 20 days. On blinded review the FNR of MRI compared to CT for multifocality was 0% vs. 38%, 50% vs 80% for satellitosis, and 22% vs 46% for MTF as a whole. CONCLUSION: CT is inadequate for pre-operative diagnosis of MTF in resectable ICC, even when performed within 30 days of hepatectomy. We recommend liver-protocol MRI as the standard pre-operative imaging modality in non-metastatic ICC.