The Effect of Histopathological Growth Patterns of Colorectal Liver Metastases on the Survival Benefit of Adjuvant Hepatic Arterial Infusion Pump Chemotherapy

BACKGROUND

Histopathological growth patterns (HGPs) are a prognostic biomarker in colorectal liver metastases (CRLM). Desmoplastic HGP (dHGP) is associated with liver-only recurrence and superior overall survival (OS), while non-dHGP is associated with multi-organ recurrence and inferior OS. This study investigated the predictive value of HGPs for adjuvant hepatic arterial infusion pump (HAIP) chemotherapy in CRLM.

METHODS

Patients undergoing resection of CRLM and perioperative systemic chemotherapy in two centers were included. Survival outcomes and the predictive value of HAIP versus no HAIP per HGP group were evaluated through Kaplan-Meier and Cox regression methods, respectively.

RESULTS

We included 1233 patients. In the dHGP group (n = 291, 24%), HAIP chemotherapy was administered in 75 patients (26%). In the non-dHGP group (n = 942, 76%), HAIP chemotherapy was administered in 247 patients (26%). dHGP was associated with improved overall survival (OS, HR 0.49, 95% CI 0.32-0.73, p < 0.001). HAIP chemotherapy was associated with improved OS (HR 0.61, 95% CI 0.45-0.82, p < 0.001). No interaction could be demonstrated between HGP and HAIP on OS (HR 1.29, 95% CI 0.72-2.32, p = 0.40).

CONCLUSIONS

There is no evidence that HGPs of CRLM modify the survival benefit of adjuvant HAIP chemotherapy in patients with resected CRLM.

Adjuvant chemotherapy for resected intrahepatic cholangiocarcinoma confers no survival advantage

560

Background: Randomized data suggest improved survival with adjuvant chemotherapy for biliary tract cancers, but subset analyses of intrahepatic cholangiocarcinoma (ICC) show limited survival benefit. This study uses a large bi-institutional cohort of resected ICC patients to evaluate the impact of adjuvant therapy on recurrence patterns and overall survival (OS) and compares these findings to data from a national cancer registry. Methods: Patients with resected ICC were identified within a bi-institutional cohort (Duke and Memorial Sloan Kettering, 1997-2020) and the National Cancer Database (NCDB, 2010-2018). Patients were stratified by treatment with adjuvant chemotherapy (adj). Site of first recurrence was categorized as local (liver only), regional (liver and perihepatic nodes), nodal (perihepatic nodes only), distant, or mixed (both liver and distant). OS was compared with Kaplan-Meier methods. Results: 367 patients underwent resection for ICC, and 163 (44%) patients received adjuvant therapy. Median follow-up was 33 vs. 44 months (adj vs observation (obs), p=0.15). 263 (72%) patients had recurrent disease, most commonly in the liver (72%). There was no difference in recurrence patterns stratified by treatment with adjuvant chemotherapy (% recurrence, adj vs obs; local: 42 vs 42; regional: 2 vs 2; nodal: 0 vs 3; distant only: 27 vs 26; mixed: 29 vs 27, p=0.5). OS was the same between groups (adj vs obs; 42 vs 49 months, p=0.3) and when stratified by recurrence site (p=0.5). Similarly, in an NCDB cohort of 1,159 ICC patients over the same time period, there was no association between adjuvant therapy and OS (adj vs obs; 49 vs 57 months, p=0.1). Conclusions: In this retrospective dual registry analysis, corroborated by national data, adjuvant chemotherapy was not associated with an improvement in OS in ICC patients subjected to curative intent resection. Further, adjuvant therapy had no impact on the high rate of hepatic recurrence, suggesting that alternative strategies, such as liver directed therapies, are needed to improve recurrence rates and OS.

Phase II trial of maximal ablative irradiation because of encasement (MAIBE) for patients with potentially resectable locally advanced pancreatic cancer

710

Background: For patients with localized but not immediately resectable pancreatic adenocarcinoma (PDAC), the role for local therapy remains undefined. Phase II MAIBE trial studied ablative radiation (A-RT) followed by consideration of surgery for patients with locally advanced pancreatic cancer (LAPC) who remain unresectable after induction chemotherapy. Methods: Participants with histologically confirmed PDAC judged unresectable by multidisciplinary review using NCCN definition after completing 3-6 months of mFOLFIRINOX (FFX) or Gemcitabine/Nab-paclitaxel (GN) were eligible. They received hypofractionated A-RT (either 67.5Gy in 15 fractions or 75Gy in 25 fractions based on anatomy) with concurrent capecitabine followed by consideration of resection within 1-3 months. Primary endpoints included resectability (80% power to detect resectability improvement from 15% in historical controls to 30% with α = 0.05) and overall survival (OS) from A-RT. Secondary endpoints included safety of surgical resection after ablative RT using 90-day Clavien-Dindo Classification of adverse events (AE). Results: Between 6/2018 and 4/2022, 47 eligible participants underwent A-RT. Median age was 67 (range, 50-80) years, 24 (51%) were male with a median tumor size of 3.95 (1.6 – 8.3) cm and CA19-9 of 92 ( < 1-1601) U/mL. Forty-four patients (94%) received at least 1 cycle of FFX with a median duration of chemotherapy (FFX or GN) of 3.5 months (1.0 – 9.4). Sixteen (34%) underwent a laparoscopy and 12 (26%) underwent a resection (Pancreaticoduodenectomy, N = 11; distal pancreatectomy, N = 1) at a median time of 3.2 months (1.9-16.9 months) from start of A-RT. The rate of resection satisfied our prespecified boundary of 11. R0 rate was 58.3%. Two-year OS from A-RT for the entire cohort was 38.9% (95% CI, 21.9 – 55.6%), including 37.1% (18.5 - 55.8%) in non-surgical and 39.4% (7.0- 72.1%) in surgical groups. There were no deaths within 90 days of surgery and 9 surgical AEs were recorded in 6 participants, including grade 1 (n = 1), grade 2 (n = 5), grade 3 (n = 2) and grade unknown (n = 1). Conclusions: In patients with LAPC and no metastatic disease after 3-6 months of chemotherapy, A-RT results in a favorable rate of resection without excess surgical toxicity. Promising 2-year OS rates were noted in both resected and non-resected patients. Clinical trial information: NCT03523312 .

Genetic heterogeneity of intrahepatic cholangiocarcinoma: Implications for outcome

595

Background: Intrahepatic cholangiocarcinoma (IHC) is characterized by marked clinical heterogeneity, likely the result of multiple cells of origin and variable driver gene alterations. The hidden-genome classifier is a statistical algorithm that classifies tumors by integrating multi-level genomic features. In this study, we trained the hidden-genome classifier with extrahepatic cholangiocarcinoma (EHC), gallbladder cancer (GBC) and hepatocellular carcinoma (HCC) as extremes of a spectrum to quantify the genetic heterogeneity of IHC with a view toward improved tumor classification. Methods: An IRB approved retrospective review of patients with biopsy confirmed IHC, EHC, GBC and HCC was conducted. All tumors were subjected to MSK-IMPACT to determine the mutational profile. A two-class model was built and internally validated with the genomic data of EHC/GBC as one class and HCC as the other class. IHC tumors were analyzed in the model and classified into three groups based on their proportional genetic resemblance to EHC/GBC (Biliary Class) or HCC (HCC Class), with the remainder as Intermediate Class. The classification thresholds were 90% resemblance to EHC/GBC or HCC and were determined by the inflection point of predicted survival. The survivals of the three groups were analyzed and compared. Results: A total of 1497 patients were included: IHC (733), EHC (208), GBC (258) and HCC (298). 527 IHC tumors with complete metagenetic information were analyzed in the model, showing a continuous spectrum of alterations, ranging from Biliary Class (122 tumors), Intermediate Class (375 tumors) to HCC Class (30 tumor). The biliary-class IHC was characterized by frequent alterations of IDH1 R132C, KRAS, SMAD4, ERBB2 gain, MDM2 gain, and CKDN2A loss, while the HCC-class IHC was primarily characterized by TERT alterations. In patients with unresected IHCs, the median survival ranged from 1 year (CI 0.77, 1.5) in Biliary Class, 1.8 years (CI 1.5, 2.0) in Intermediate Class, to 2 years (CI 0.93, NR) in HCC Class. In patients subjected to resection, the median survival of Biliary Class (2.4 years, CI 2.1, NR) was lower than both the Intermediate Class (5.1 years, CI 4.8, 6.9) and the HCC Class (3.4 years, CI 2.7, NR). Conclusions: By integrating multi-level genomic features, we leveraged the mutational heterogeneity to classify IHC based on its resemblance to EHC/GBC or HCC tumors. We found that the survival in IHC patients appeared to decline with increasing genomic similarity to Biliary Class. The results support a genomic basis for IHC’s variable clinical behavior and point to a role of mutational testing to guide clinical intervention. [Table: see text]

Adjuvant modified FOLFIRINOX (mFFX) for resected pancreatic cancer (PDAC): Real world outcomes (RWO)

685

Background: Adjuvant mFFX (5-fluorouracil, leucovorin, irinotecan, oxaliplatin) is a standard-of-care for fit patients (pts) with resected PDAC, owing to the immediate practice-changing PRODIGE 24/CCTG PA6 trial (2018). Five-year follow-up: median overall survival (mOS) 53.3 months (m) and median disease-free survival (mDFS) 21.4 m for mFFX vs 35.5 m and 12.8 m for gemcitabine (Conroy, JAMA Onc, 2022). RWO for pts outside a clinical trial are lacking. Herein, we report RWO for pts with resected PDAC and intent for adjuvant mFFX at Memorial Sloan Kettering (MSK). Methods: Institutional databases were queried to identify pts with resected PDAC who received any dose of adjuvant mFFX. Demographic, clinicopathologic, genomic, dosing details, and survival data were abstracted from medical and pharmacy records. Primary endpoint was to determine recurrence-free survival (RFS) calculated from start date mFFX to disease recurrence or death and OS calculated from start date mFFX to death. Secondary endpoints included dose reductions, significant treatment delay, toxicity profile, patterns of failure, genomic associations with outcome. RFS and OS are estimated using the Kaplan-Meier method. Study approved by MSK IRB. Results: N = 114 pts with resected PDAC treated with mFFX (> 1 dose) identified between 01/2015- 01/2022. Median age: 67 years (range 35 to 82); N = 43 (38%) > 70 years, N = 18 (16%) > 75 years, N = 2 (2%) > 80 years. Baseline Performance Status recorded in N = 104: N = 31 (30%) ECOG 0, N = 64 (62%) ECOG 1, N = 9 (9%) ECOG 2. Disease stage: N = 36 (32%) stage III, N = 61 (54%) stage II, and N = 17 (15%) pts stage I. Resection status: N = 91 (80%) R0, N = 23 (20%) R1. Presence of lymphovascular invasion: N = 92 (81%), perineural invasion N = 106 (93%). Median baseline CA 19-9: 20 U/mL (IQR; 9, 38). Median follow up: 22.4 m (range 6.2, 50.4). Median time from surgery to start mFFX: 7.4 weeks (IQR; 6.1, 9.3). Median # of mFFX doses received: 12 (IQR; 12, 12), N = 90 (79%) pts completed 12 doses. Dosing details available N = 112. N = 55 (49%) prescribed less than full dose of > one drug at baseline. Dose reductions: N = 57 (51%). N = 69 (62%) received < 12 doses oxaliplatin. N = 97 (87%) received growth factor support. mRFS: 31 m (95% CI; 23, Not Reached). N = 18 (16%) were hospitalized for treatment related adverse events, no therapy related mortality. N = 24 (21%) received adjuvant radiation therapy. One-year OS rate: 93% (95% CI; 89%, 98%) and 2-year OS rate: 78% (95%CI: 70%, 88%). Among patients with recurrence (N = 44), most common sites of first recurrence were: liver (N = 18, 41%), local (N = 14, 32%), and lung (N = 9, 20%). Conclusions: These data endorse mFFX as standard therapy for resected PDAC. The survival signals are encouraging in a prognostically unfavorable albeit select patient population (relative to PRODIGE 24). Dose adjustments to facilitate optimizing tolerability is key. Additional genomic and subtype analyses are underway.

Concordance in oncogenic alterations between primary and recurrent/metastatic cholangiocarcinoma pairs using targeted next-generation sequencing

604

Background: The genetic background of cholangiocarcinoma (CCA) commonly involves alterations in kinase signaling, tumor suppression, oxidative stress modulation, and proto-oncogenic coupling pathways. Novel agents targeting such pathways have shown promise in systemic treatment; however, studies examining differences in the mutational landscapes between primary and recurrent, metastatic, or progressive disease after systemic therapy are lacking. The present study aimed to determine if recurrent, metastatic, or progressive disease genetically parallels the primary or not. Methods: Patients with biopsy proven CCA (primary tumor and paired recurrent/metastatic or progressive disease) from two institutions (MSKCC and Duke) were identified. Targeted next-generation sequencing (Integrated Mutation Profiling of Actionable Cancer Targets (IMPACT)) capturing single nucleotide variants, copy number alterations, and structural variants was used to compare driver alteration concordance across the paired samples. Subgroup analyses were performed based on exposure to systemic therapy in patients with disease progression and tumor type (intrahepatic versus extrahepatic). Results: Sample pairs from 65 patients with intrahepatic (ICCA, n=54) and extrahepatic CCA (ECCA, n=11) were analyzed. Median time between samples was 19.6 months (range 2.7 - 122.9). Some de novo alterations were identified in recurrent/metastatic samples, but overall concordance (70%) was demonstrated between patient pairs for common oncogenic driver genes (Table). Subgroup analyses of summative ICCA and ECCA mutations revealed concordance of 65% and 88%, respectively. Concordance was also demonstrated between pairs exposed to systemic therapy between sample collections (n=50, 71%). Conclusions: In this dataset of CCA patients, a concordance rate of 70% was identified in the genomic alterations between primary and recurrent/metastatic pairs, and this did not appear to be altered by prior treatment with systemic chemotherapy. While limited by sample size, concordance in ICCA pairs was lower than that seen in ECCA. [Table: see text]

Phase I trial of adjuvant autogene cevumeran, an individualized mRNA neoantigen vaccine, for pancreatic ductal adenocarcinoma

2516

Background: Pancreas ductal adenocarcinoma (PDAC) is a lethal cancer that claims ̃90% of patients in <24 months of diagnosis. PDAC is also refractory to immunotherapy as most tumors exhibit an immune excluded/desert phenotype. However, although characterized by low mutation rates, most PDACs harbor mutations that can generate immunogenic neoantigens. Here, we report the results of a phase-I trial of autogene cevumeran, a systemic RNA-lipoplex individualized neoantigen-specific immunotherapy (iNeST) vaccine, to stimulate immunity against neoantigens in resected PDAC patients. Methods: We conducted an investigator-initiated, single-center, phase-I trial of adjuvant autogene cevumeran containing up to 20 neoantigens in each individualized vaccine, identified from resected PDACs using real-time next generation sequencing and bioinformatic neoantigen discovery. Following surgery, patients received atezolizumab (1 dose; week 6), autogene cevumeran (8 weekly doses starting week 9; doses 9,10 – weeks 17, 46), and modified (m) FOLFIRINOX (12 cycles; starting week 21). Primary endpoint: safety. Other endpoints: feasibility (actual vs. target treatment time), vaccine response (responder = positivity by two independent blood assays: IFNg ELISpot and T cell clonal expansion), and recurrence-free survival (RFS). Target accrual: n=20. Results: n=19 patients underwent surgery and received atezolizumab at 6.3 weeks (median; 95% CI 6.0–6.57) after surgery with no ≥ grade 3 (Gr3) adverse events. n=16/19 patients (84%) received autogene cevumeran at 9.4 weeks (median; 95% CI 9–10) after surgery. n=1/19 (5%) had insufficient neoantigens for vaccine manufacture. n=1/16 (6%) developed a vaccine-related Gr3 fever and hypertension. n=15/16 vaccinated patients (94%) received mFOLFIRINOX (median 12 cycles; 95% CI 7–12). Autogene cevumeran expanded polyclonal (median 7.5 clones, 95% CI 2–28), IFNg-producing neoantigen-specific CD8+ T cells in 50% (n=8/16) of patients from undetectable levels to large fractions (median 2.9%, Table) of all blood T cells. At an early median follow-up of 15 months, vaccine responders (n=8) had a longer RFS vs. non-responders (n=8) (median not reached vs. 13.7 months, HR 0.08, 95% CI 0.01-0.5, P = 0.007). Conclusions: Autogene cevumeran is safe, feasibly manufactured in a clinically relevant timeframe, and immunogenic in PDAC. Vaccine induced neoantigen-specific immunity preliminarily correlates with improved PDAC outcome. Further clinical trials in PDAC are warranted. (This imCORE Network project was funded by Genentech Inc and BioNTech; additional funding from Stand Up To Cancer, Lustgarten Foundation). Clinical trial information: NCT04161755. [Table: see text]

Radiation segmentectomy of hepatic metastases with Y-90 glass microspheres

PURPOSE

To evaluate safety and efficacy of radiation segmentectomy (RS) with ⁹⁰Y glass microspheres in patients with limited metastatic liver disease not amenable to resection or percutaneous ablation.

METHODS

Patients with ≤ 3 tumors treated with RS from 6/2015 to 12/2017 were included. Target tumor radiation dose was > 190 Gy based on medical internal radiation dose (MIRD) dosimetry. Tumor response, local tumor progression (LTP), LTP-free survival (LTPFS) and disease progression rate in the treated segment were defined using Choi and RECIST 1.1 criteria. Toxicities were evaluated using modified SIR criteria.

RESULTS

Ten patients with 14 tumors underwent 12 RS. Median tumor size was 3 cm (range 1.4-5.6). Median follow-up was 17.8 months (range 1.6-37.3). Response rates per Choi and RECIST 1.1 criteria were 8/8 (100%) and 4/9 (44%), respectively. Overall LTP rate was 3/14 (21%) during the study period. One-, two- and three-year LTPFS was 83%, 83% and 69%, respectively. Median LTPFS was not reached. Disease progression rate in the treated segment was 6/18 (33%). Median overall survival was 41.5 months (IQR 16.7-41.5). Median delivered tumor radiation dose was 293 Gy (range 163-1303). One major complication was recorded in a patient post-Whipple procedure who suffered anaphylactic reaction to prophylactic cefotetan and liver abscess in RS region 6.5 months post-RS. All patients were alive on last follow-up.

CONCLUSION

RS of ≤ 3 hepatic segments can safely provide a 2-year local tumor control rate of 83% in selected patients with limited metastatic liver disease and limited treatment options. Optimal dosimetry methodology requires further investigation.

Immediate post-thermal ablation biopsy of colorectal liver metastases to predict oncologic outcomes

4602

Background: Thermal ablation (TA) is used as a local cure for selected colorectal liver metastases (CLM) with minimal risk. A critical limitation of TA has been early local tumor progression (LTP). The goal of this study is to establish the role of ablation zone (AZ) biopsy in predicting LTP. Methods: This institutional review board-approved prospective study included patients with CLM of 5cm or less in maximum diameter, with confined liver disease or stable, limited extrahepatic disease. Both radiofrequency(RF) and microwave(MW) ablation modalities were used. A biopsy of the center and margin of the AZ was performed immediately after ablation. The applicators were also examined for the presence of viable tumor cells. All samples containing morphologically identified tumor cells were further interrogated with immunohistochemistry to determine the proliferative and viability potential of the detected tumor cells. Ablation margin size was evaluated on the first CT scan performed 4–8 weeks after ablation and was confirmed by 3D assessment with Ablation Confirmation Software (Neuwave™). Variables were evaluated as predictors of time to LTP with the competing-risks model (uni- and multivariate analyses). Results: Between November 2009 and February 2019, 102 patients with 182 CLMs were enrolled. Mean tumor size was 2.0 cm (range, 0.6–4.8 cm). MW was used in 95/182 (52%) tumors and RF in 87/182 (48%). Median follow-up was 19 months. Technical effectiveness was evident in 178/182 (97%) ablated tumors on the first contrast material–enhanced CT at 4–8-weeks post-ablation. The cumulative incidence of LTP at 12 months was 19% (95% confidence interval [CI]: 14, 27). Samples from 64 (35%) of the 178 technically successful cases contained viable tumor. At univariate analysis, tumor size, minimal margin size, and biopsy results were significant in predicting LTP. In a multivariate model, margin size of less than 5 mm (P < .001; hazard ratio [HR], 4.3), and positive biopsy results (P = .02; HR, 1.8) remained significant. LTP within 12 months after TA was noted in 3% (95% CI: 1, 6) of tumor-negative biopsy CLMs with margins of at least 5 mm. Conclusions: Biopsy and pathologic examination of the AZ predicts LTP regardless of TA modality used. This can optimize ablation as a potential local cure for patients with limited CLM.

Hepatic arterial infusion (HAI) chemotherapy via the medtronic pump-assessing dose delivery, response rates and toxicity

e16007

Background: Due to the termination of the Codman pump, in order to administer hepatic arterial infusion (HAI) chemotherapy, the Medtronic pump has been used with the Codman catheter at MSKCC since 2018. Methods: Retrospective review of patients(pts) receiving HAI therapy via Medtronic pumps. Expected versus actual dose delivery of HAI FUDR, response rates and safety were reviewed. Results: Pts included; unresectable colorectal liver metastases (CRLM) (94 pts), resected CLRM (66 pts) and unresectable intrahepatic cholangiocarcinoma (ICC) (11 pts). Baseline characteristics of the 171 evaluable pts are shown in Table. In 120 pts with mCRC, 51.8% (n = 58) had 10 or more hepatic tumors and 23.2%( n = 26) had 50% or more liver involvement by metastases. 52 out of 171 pts (30.4%) had 100% of the expected FUDR doses by completion of the 3rdcycle. Another 49.7% (85/171) had greater than or equal to 50% of the expected FUDR dose. 96 pts had measurable disease (unresectable CLRM and ICC subgroups) evaluated by RECIST 1.1 with an MSK radiologist. The partial response for unresectable CRLM pts was 47% (41/87), and 28% (24/87) were stable. The partial response for ICC pts was 22% (2/9), and 33% (3/9) were stable. Conclusions: We evaluated 171 pts who were heavily pretreated and had extensive disease prior to pump placement. Partial response by RECIST 1.1 of 47% was evident in those pts with CRLM and 22% for pts with ICC. Dose delivery of HAI FUDR was compatible with that seen with the Codman pump and no increase in toxicity was noted. An updated analysis with additional evaluable pts will be presented at the meeting. [Table: see text]

Randomized phase II trial of adjuvant hepatic arterial infusion (HAI) + systemic FOLFIRI +/- panitumumab (Pmab) in patients with resected RAS wild type colorectal cancer hepatic metastases (CRLM)

3555

Background: HAI therapy has improved recurrence free (RF) survival in several randomized trials after resection of CRLM. The purpose of this trial was to determine whether systemic Pmab added to adjuvant HAI + FOLFIRI in RAS WT pts increases 15 months RF survival (RFS) after liver resection. Secondary endpoints are evaluation of overall survival, toxicity, and predictive biomarkers. Methods: RAS WT pts with resected liver mets were randomized to HAI + SYS (+/-) Pmab after stratification by clinical risk score (≥ 3 or < 3) and previous chemotherapy (Y/N). For a particular arm, if 24 or more patients (pts) were alive and RF at 15 months, the regimen in that arm would be considered worthy of further investigation. The initial dose of HAI FUDR 0.12mg/Kg + dexamethasone was infused over the first 2 weeks of a 5-week cycle. Systemic chemotherapy was delivered on days 15 and 29 (irinotecan 125 mg/m2, LV 400 mg/ m2, 5FU 1000 mg/m2 48-hour continuous infusion and +/- Pmab 6mg/kg). Patient characteristics were compared between arms using Fisher’s exact test and Wilcoxon rank-sum test. Survival curves were estimated using the Kaplan-Meier method and compared by the log-rank test. Results: After randomization of 75 pts, the arm receiving + Pmab met the decision rule of having ≥ 24 patients alive and RF at 15 months. The two arms had similar pt characteristics and toxicity, with the exception of Pmab related rash (Table). The 15-month RFS is 79% and 67% in +/- Pmab arms, respectively. With a median follow-up of 45 months, 3-year RFS is 65% [CL 0.45-0.78] and 42% [CL 0.24-0.57], and 3-year survival is 96% and 90% in +/-Pmab arms, respectively. Conclusions: In this trial, the addition of Pmab to HAI and SYS showed promising activity without increase in biliary toxicity and should be further investigated in a larger study. Predictive biomarkers will be presented. Clinical trial information: NCT01312857. [Table: see text]

Adjuvant hepatic arterial infusion pump chemotherapy and resection versus resection alone in patients with low-risk resectable colorectal liver metastases - the multicenter randomized controlled PUMP trial

Background

Recurrences are reported in 70% of all patients after resection of colorectal liver metastases (CRLM), in which half are confined to the liver. Adjuvant hepatic arterial infusion pump (HAIP) chemotherapy aims to reduce the risk of intrahepatic recurrence. A large retrospective propensity score analysis demonstrated that HAIP chemotherapy is particularly effective in patients with low-risk oncological features. The aim of this randomized controlled trial (RCT) --the PUMP trial-- is to investigate the efficacy of adjuvant HAIP chemotherapy in low-risk patients with resectable CRLM.

Methods

This is an open label multicenter RCT. A total of 230 patients with resectable CRLM without extrahepatic disease will be included. Only patients with a clinical risk score (CRS) of 0 to 2 are eligible, meaning: patients are allowed to have no more than two out of five poor prognostic factors (disease-free interval less than 12 months, node-positive colorectal cancer, more than 1 CRLM, largest CRLM more than 5 cm in diameter, serum Carcinoembryonic Antigen above 200 μg/L). Patients randomized to arm A undergo complete resection of CRLM without any adjuvant treatment, which is the standard of care in the Netherlands. Patients in arm B receive an implantable pump at the time of CRLM resection and start adjuvant HAIP chemotherapy 4–12 weeks after surgery, with 6 cycles of floxuridine scheduled. The primary endpoint is progression-free survival (PFS). Secondary endpoints include overall survival, hepatic PFS, safety, quality of life, and cost-effectiveness. Pharmacokinetics of intra-arterial administration of floxuridine will be investigated as well as predictive biomarkers for the efficacy of HAIP chemotherapy. In a side study, the accuracy of CT angiography will be compared to radionuclide scintigraphy to detect extrahepatic perfusion. We hypothesize that adjuvant HAIP chemotherapy leads to improved survival, improved quality of life, and a reduction of costs, compared to resection alone.

Discussion

If this PUMP trial demonstrates that adjuvant HAIP chemotherapy improves survival in low-risk patients, this treatment approach may be implemented in the standard of care of patients with resected CRLM since adjuvant systemic chemotherapy alone has not improved survival.

Trial registration

The PUMP trial is registered in the Netherlands Trial Register (NTR), number: 7493. Date of registration September 23, 2018.

A retrospective study of hepatic arterial infusion (HAI) FUDR/Dex and mitomycin C (MMC) for chemotherapy refractory unresectable intrahepatic cholangiocarcinomas (ICC)

432

Background: ICC are aggressive tumors with approximately 6,000 cases a year in US. The 5-year survival rate is less than 30% even for localized disease. There is only one approved line of systemic (SYS) treatment and further treatment options are necessary. HAI chemotherapy is an option to treat liver predominant cancers. Methods: After obtaining IRB approval, we retrospectively reviewed patients (pts) with ICC chemo refractory unresectable liver limited (LL) or liver dominant (LD) disease who received intrahepatic chemotherapy with HAI MMC. Baseline characteristics, previous lines of therapy, toxicity profile, combinations and radiographic responses were reviewed. Tumor genomic analyses were performed on samples using an on-site next generation sequencing (NGS) assay. Results: Between January 2011 and October 2018, 19 patients ICC with LL or LD disease were treated with HAI FUDR/Dex/MMC at Memorial Sloan Kettering Cancer Center. Disease was confined to the liver in 58% of the pts. All pts had previous chemotherapy (1-4 lines) and 14 (74%) previously had HAI FUDR/Dex. Of the 19 pts, 56% had HAI with FUDR/Dex and MMC, 43% had FUDR/Dex, MCC and SYS and 5% had HAI MMC and SYS. Seventeen patients were evaluable for response, two are being treated and will have response assessment for the meeting. Response was noted in 4 (23.5%), stable disease in 6 (35.5%) and progressive disease in 7 (41%) pts. Median overall survival from treatment was 6.1months (0.36-26). Median progression free survival was 3.65 months (0.36-9.53). Four patients had dose reductions. Common toxicity attributed to MMC was grade (G) one fatigue (32%), thrombocytopenia G1(16%) and G2 (5%). Of the 12 tumors analyzed to date the most 92% of tumors harbored at least one (0-10) genomic alteration. Common genomic alterations were ARID1 (25%), RASA1 (25%), IDH1(16.6%), NTRK (16.6%), TERT (16.6%), NRAS (16.6%), CDKN2 (16. 6%). FGFR2-FOXP1 and GTL2MEt fusions were found in one patient each. Conclusions: HAI FUDR/Dex/MMC containing regimens are active in pts with heavily pretreated refractory unresectable ICC. This strategy should be further investigated. Translational data will be presented.

Genetic And Morphological Evaluation (GAME) score for patients with colorectal liver metastases

BACKGROUND

This study sought to develop a clinical risk score for resectable colorectal liver metastasis (CRLM) by combining clinicopathological and clinically available biological indicators, including KRAS.

METHODS

A cohort of patients who underwent resection for CRLM at the Johns Hopkins Hospital (JHH) was analysed to identify independent predictors of overall survival (OS) that can be assessed before operation; these factors were combined into the Genetic And Morphological Evaluation (GAME) score. The score was compared with the current standard (Fong score) and validated in an external cohort of patients from the Memorial Sloan Kettering Cancer Center (MSKCC).

RESULTS

Six preoperative predictors of worse OS were identified on multivariable Cox regression analysis in the JHH cohort (502 patients). The GAME score was calculated by allocating points to each patient according to the presence of these predictive factors: KRAS-mutated tumours (1 point); carcinoembryonic antigen level 20 ng/ml or more (1 point), primary tumour lymph node metastasis (1 point); Tumour Burden Score between 3 and 8 (1 point) or 9 and over (2 points); and extrahepatic disease (2 points). The high-risk group in the JHH cohort (GAME score at least 4 points) had a 5-year OS rate of 11 per cent, compared with 73·4 per cent for those in the low-risk group (score 0-1 point). Importantly, in cohorts from both the JHH and MSKCC (747 patients), the discriminatory capacity of the GAME score was superior to that of the Fong score, as demonstrated by the C-index and the Akaike information criterion.

CONCLUSION

The GAME score is a preoperative prognostic tool that can be used to inform treatment selection.

Association of adjuvant hepatic artery infusion chemotherapy after resection of colorectal liver metastases with improved survival in patients with both right- and left-sided primary tumors

776

Background: Patients with a right-sided (R) and/or KRAS-mutated (KRAS-MUT) primary tumors have inferior outcomes compared to patients with left sided (L) and/or KRAS-wildtype (KRAS-WT) cancers. Adjuvant hepatic artery pump infusion (HAI) chemotherapy improves overall survival (OS) in patients with colorectal liver metastases (CRLM). We investigated the impact of HAI in relation to the laterality and KRAS status of the primary tumors. Methods: Patients with resected CRLM and available KRAS-status, treated with and without adjuvant HAI were reviewed from a prospective institutional database. Rectal tumors were excluded. Correlations between adjuvant HAI, clinicopathological factors including laterality, KRAS status and OS were analyzed. Cox proportional hazard regression was used to assess survival outcomes comparing R (cecum to transverse colon, excluding appendix) versus L (splenic flexure to sigmoid) colon cancers. Results: 490 patients (R, n = 183; L, n = 307) were evaluated between 1993-2012 (median follow up of 6.5 years). Fifty-six percent (n = 277) received adjuvant HAI. Adjuvant HAI was associated with improved median OS in both R (7.0 vs. 4.4 years, p = 0.006) and L tumors (10.5 vs. 5.4 years, p < 0.01). On multivariate analysis, HAI remained associated with improved OS (HR 4.49, p = 0.001) independent of primary tumor site and other clinical predictors (Table). Conclusions: Adjuvant HAI after resection of CRLM is independently associated with improved OS regardless of laterality of primary tumor. Treatment with adjuvant HAI correlates with improved prognosis in patients with resectable L/KRAS-MUT and R/KRAS-WT tumors. The biological difference for these outcomes requires further investigation. [Table: see text]

Extraordinary survivorship after colorectal liver metastasis resection to identify a distinct molecular profile associated with survival in an independent cohort of 965 patients

3581

Background: Half of patients diagnosed with colorectal cancer (CRC) develop metastases and most are CRC liver metastasis (CRLM). A mere 20% of these patients undergo complete resection of their liver disease and 5-year overall survival (OS) is only 50%. We hypothesized that we could identify a specific molecular profile associated with extraordinary survivorship in CRLM patients that would more precisely inform underlying tumor biology beyond standard clinical and pathologic features. Methods: Tumor samples were identified from patients who underwent curative resection. Patients with disease-specific survival (DSS) ≥10 years following resection were compared to those with ≤2-year survival (10yr vs. 2yr). Evaluable DNA was obtained from 36 cases (2yr, n = 17; 10yr, n = 19) then sequenced and analyzed with MSK-IMPACT (MSK-I), a hybridization capture, next generation sequencing platform. Differentially altered genes in 10yr vs. 2yr cohorts were identified (Fisher’s exact). Findings in the extraordinary survivors group were validated using MSK-I in an independent cohort of 965 metastatic CRCs (metCRCs). Kaplan-Meier estimates and log-rank test were used. Results: In the 2yr group, we noted higher clinical risk scores and more complex chemotherapy regimens vs. the 10yr group. Molecularly, mutually exclusive KRAS and TP53 mutations were noted in the 10yr group, whereas significant co-occurrence of KRAS and TP53 mutations was seen in the 2yr group. Further, we noted significant enrichment of VEGF copy number gains in the 2yr group vs. the 10yr group. APC mutation was equally common. In the validation cohort, KRASmut/TP53wt and TP53mut/KRASwt patients (median OS of 10 and 15 years respectively) had significantly better OS than the co-occurring KRASmut / TP53mutpatients (median OS of 4.9 years; (P = 0.0001)). Conclusions: Single mutation of either KRAS or TP53 is associated with better outcomes than co-occurring KRAS/TP53 mutations in metCRC. These data demonstrate use of an extraordinary survivor cohort to identify a molecular profile associated with significant survival differences in an independent cohort of metCRC patients.

Mutational profiling of resected intrahepatic cholangiocarcinoma

e15675

Background: Intrahepatic cholangiocarcinoma (ICC) is an aggressive neoplasm with increasing incidence and mortality. Resection is the only potential curative treatment and is associated with 5-year survival up to 44%. The objective of this study was to characterize the mutational landscape of patients with ICC undergoing resection and to identify potential prognostic genetic markers that may be unique to these patients. Methods: Sixty-six resected ICC tumor specimens were assessed for genetic alterations using next-generation sequencing of 410 cancer genes by Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT). We identified the gene alterations occurring with greatest frequency and grouped mutations by known cancer pathways and families, such as RAS-MAPK pathway, mTOR pathway, Notch signaling pathway, chromatin-remodeling gene family, and DNA repair gene family. Overall survival (OS) was calculated from time of resection until death and disease-free survival (DFS) was calculated from resection until recurrence or death. Kaplan Meier 5-year estimates and the log-rank test were used to evaluate the associations with OS and DFS, adjusted for multiple comparisons with false discovery rate (FDR) correction. Results: The median age of all patients was 65 years (range 29-87 years). Majority of tumors were T1 (24/66, 36%) or T2 (35/66, 53%), and of moderate differentiation (46/66, 70%). Lymphovascular invasion, perineural invasion, and periductal infiltration were present in 34 (52%), 20 (30%), and 8 (12%) patients, respectively. The median number of genetic alterations per tumor was 3 (range 0-26). The most common genetic alterations were PBRM1 (16/66, 24%), IDH1 (15/66, 23%), ARID1A (14/66, 21%) and TP53 (8/66, 12%). FGFR2 fusion mutations (5/66, 8%) were relatively rare. The median OS for all patients was 53.4 months (95%CI: 43.0-79.3 months) and median DFS was 17.4 months (95%CI: 10.4-32.6 months). None of the gene alterations or pathways were associated with OS (p = 0.29-0.84) or DFS (p = 0.23-0.65). Conclusions: In this cohort of resected ICC patients, genetic alterations or alterations within gene families, by themselves, did not stratify risk of disease recurrence or death.

Poor prognosis in patients with resectable KRAS-MUT colorectal liver metastases is improved by adjuvant hepatic arterial infusion chemotherapy

e15161

Background: Patients with KRAS-mutated colorectal cancer liver metastases (CRLM) have worse outcomes after resection. Adjuvant hepatic artery pump chemotherapy (HAIC) improves overall survival. We investigated the impact of HAIC in relation to KRAS mutational status in patients with resected CRLM. Methods: Patients with resected CRLM treated with and without adjuvant HAIC and available KRAS status (wild-type, WT; mutated, MUT) were reviewed from a prospective institutional database. Correlations between KRAS status, adjuvant HAIC, clinical factors, and overall survival (OS) were analyzed. Cox proportional hazard model was used to adjust for confounders. Results: Between 1993-2012, 675 patients (419 KRAS-WT, 256 KRAS-MUT) with a median follow up of 6.5 years after resection were evaluated. Fifty-four percent received adjuvant HAIC. Tumor characteristics (synchronous disease, number of lesions, clinical-risk score, 2-stage hepatectomy) were significantly worse in the HAIC group, however, there were more patients with extrahepatic metastases in the no-HAIC group. Adjuvant HAIC was associated with improved OS in both KRAS-WT (5-yr OS 76% vs 57%, HR 0.51, p<0.01) and KRAS-MUT (5-yr OS 59% vs 40%, HR 0.56, p<0.01) patients. On multivariate analysis, HAIC remained associated with improved OS (HR 0.6, p<0.001) independent of KRAS status and other clinico-pathologic factors. Conclusions: Adjuvant HAIC after resection of CRLM is independently associated with improved OS regardless of KRAS status and may abrogate the poor prognosis in resectable KRAS-MUT CRLM. [Table: see text]

Right versus left: Impact of primary location on survival and cure in patients undergoing hepatic resection for metastatic colon cancer

664

Background: Recent prospective studies in metastatic colorectal cancer (mCRC) have demonstrated an association between left-sided primaries and improved overall (OS) and progression-free survival (PFS). Primary location (right vs left colon) has not been well studied in patients undergoing potentially curative hepatic resection. Methods: A single-institution database was queried for all initial hepatic resections for mCRC 1992-2004. Postoperative deaths and patients with no followup after 90 days were excluded. Primary location determined by chart review (Right = cecum to transverse; Left = splenic flexure to sigmoid). Rectal cancer (distal 16cm), multiple primaries, and unknown location were excluded. Kaplan Meier and Cox regression methods were used. Cure was defined as actual 10-year survival with no recurrence or resected recurrence with at least 3 years of disease-free followup. Results: 907 patients were included with a median followup of 11 years. 578 patients (64%) had left-sided and 329 (36%) had right-sided primary. Median OS for patients with a left-sided primary was 5.2 years (95% CI: 4.6-6.0) versus 3.6 years (95% CI: 3.2-4.2) for right-sided (p = 0.004). The hazard ratio (HR) for right-sided tumors was 1.22 (95% CI: 1.02-1.45, p = 0.028) after adjusting for age, CEA > 200, DFI < 12 months, hepatic tumor > 5cm, > 1 tumor, lymph node status, margin, and extrahepatic disease. Recurrence-free survival (RFS) was marginally different stratified by primary location (p = 0.065). Estimated cure rates were 22% for left and 20% for right-sided tumors. Conclusions: Among patients selected for hepatic resection of metastatic colon cancer, left-sided primary tumors were associated with an improved OS but not RFS. This difference in OS was independent of common prognostic variables. Estimated cure rates were not statistically different. Patients with left-sided primary tumors display a prolonged clinical course after recurrence suggestive of more indolent biology. [Table: see text]

Hepatic artery therapies for unresectable colorectal liver metastases: Pooled survival analysis of 968 patients from TACE, yttrium-90, and HAI studies

747

Background: Approximately 140,000 people are newly diagnosed with colorectal cancer (CRC) annually, while 50,000 will die from the disease. Median survival has increased with use of modern chemotherapy and biologic agents. Sixty percent of advanced CRC patients (pts) present with or develop liver metastases, with additional treatment options such as hepatic artery therapies. However, there is no strong scientific evidence to support the use of transarterial chemoembolization (TACE), Yttrium 90 (Y90) or chemotherapy infusion (HAI). Methods: Pooled survival analysis of 968 pts with unresectable colorectal liver metastases (CLRM) treated with hepatic artery therapies, including HAI, Y90 or TACE. A rigorous selection of studies was conducted to ensure exchangeability across studies and adequate comparison of groups. Kaplan Meier survival curves were reconstructed from original publications. Data was further stratified into presence or absence of extra-hepatic disease (EHD). Results: Total of 968 pts with mean age 60.5 years, 65.5% male, 64.9% ECOG 0 and 42.4% with EHD. Patient characteristics in the various studies as per Table. Initial analysis showed a significant reduction in the hazard of death comparing HAI & Y90 (HR = 0.45[0.38, 0.55]). There was a 67% reduction in the hazard of death comparing HAI to TACE (HR = 0.33[0.27, 0.41]). Median survival in pts without and with EHD was 32 &16 months for HAI, 11 & 6 months for Y90 and 11 & 8 months for TACE (p < 0.0001). Conclusions: Based on our analysis, HAI offers better survival rates when compared to TACE and Y90. This survival advantage for HAI is not only limited to CRC pts with hepatic disease only, but also is evident in those pts with unresectable CLRM in the setting of co-existent EHD. [Table: see text]

Two-stage hepatectomy for colorectal liver metastases: A multi-institutional retrospective review

351

Background: A significant number of patients with colorectal liver metastases (CRLM) present with unresectable bilobar disease. Two-stage hepatectomy with portal vein embolization (PVE) have been described as a treatment for CRLM allowing for volume regeneration of a functional liver remnant. No large-scale multi-institutional studies exist. The aim of this project was to describe outcomes following two-stage hepatectomy, including overall survival (OS), recurrence-free survival (RFS), and complications. Methods: Patients completing two-stage hepatectomy for CRLM at five US institutions were identified and retrospectively analyzed (2000-2015). Overall survival and recurrence-free survival following second-stage surgery, short-term mortality, Clavien-Dindo complications, and readmission rates were examined. Results: A total of 209 patients were identified. Mean age was 52 (SD +/-11.4), 59.8% were male, and 87.0% had synchronous disease. A total of 65.1% of patients underwent PVE, and 27.3% underwent hepatic artery infusion pump placement. 88.3% of cases underwent neoadjuvant chemotherapy. Following the first stage, 30-day morbidity was 24.4%, with 4.8% major (Clavien-Dindo grade ≥3) complications, and 30-day readmission was 6.7%. Mean time between first and second stage was 4.1 months (S.D. +/-3.1), and 57.5% received systemic chemotherapy between the two resections. Following the second stage, overall complications were 47.4% with 23.9% major complications, and 30-day readmission was 9.7%. Mortality following second stage was 3.8% at 30 days, and 5.3% at 90-days. Following the second stage, RFS at 1-, 3-, and 5-years was 80%, 46%, and 29% respectively. OS at 1-, 3-, and 5-years was 87%, 64%, and 45% respectively. Conclusions: Two-stage hepatectomy for CRLM provides acceptable recurrence-free and overall survival in the context of advanced bilobar disease. Major complications and readmission following the first stage are rare. Following the second stage, short-term major morbidity, mortality, and readmissions are also acceptable. For well-selected patients, two-stage hepatectomy remains a safe and effective treatment for CRLM, with potential for more widespread adoption.

Hepatic adenoma among adult survivors of childhood cancer

117

Background: Hepatic adenoma is a rare and poorly understood benign epithelial neoplasm. Because of the potential for spontaneous hemorrhage, rupture, or malignant transformation; hepatic adenoma over 5 cm require resection. In the general population, the prevalence of hepatic adenoma is estimated at 1 in 100,000 and identified predominantly in obese females on oral contraceptives. An increased risk for hepatic adenoma among adult survivors of childhood and young adult cancer has not been previously reported Methods: Cancer diagnosis and treatment, as well as demographic factors, medications, and comorbidities, were collected from the medical chart among patients with pathological confirmation of hepatic adenoma. All cases were patients diagnosed with a non-hepatic cancer before the age of 40 and seen at Memorial Sloan Kettering Cancer Center. Results: Twelve cases of hepatic adenoma were pathologically confirmed; seven patients (58%) had more than one adenoma. Eleven (92%) cases were female. The most common preceding cancer diagnosis was leukemia (N = 4; 33%). Five (42%) had undergone allogeneic hematopoietic cell transplant with total body irradiation (TBI) as part of their preconditioning regimen. Cases were not as a rule obese; median body mass index was 22.2 kg/m2 (range, 17.6-31.0 kg/m2). All eleven females had a history of current or prior hormone therapy with estrogen and progesterone; the single male case was hypogonadal as a result of radiation therapy to the testes during treatment for acute myelogenous leukemia (AML) and was receiving testosterone therapy at the time of chart review. Eight patients (67%) had hypothyroidism and two (17%) were taking anti-epileptic drugs. Only two patients (17%) were monitored radiographically following biopsy; seven patients (58%) underwent hepatic resection and three (25%) underwent embolization. No patient had significant blood loss or has been observed to undergo malignant transformation, although follow-up is ongoing. Conclusions: Adult survivors of childhood and young adult cancer, particularly females with a history of current or prior hormone therapy, may be at increased risk for hepatic adenoma. Further investigation of this potentially morbid condition is warranted.

Survival after resection of perihilar cholangiocarcinoma-development and external validation of a prognostic nomogram

BACKGROUND

The objective of this study was to derive and validate a prognostic nomogram to predict disease-specific survival (DSS) after a curative intent resection of perihilar cholangiocarcinoma (PHC).

PATIENTS AND METHODS

A nomogram was developed from 173 patients treated at Memorial Sloan Kettering Cancer Center (MSKCC), New York, USA. The nomogram was externally validated in 133 patients treated at the Academic Medical Center (AMC), Amsterdam, The Netherlands. Prognostic accuracy was assessed with concordance estimates and calibration, and compared with the American Joint Committee on Cancer (AJCC) staging system. The nomogram will be available as web-based calculator at mskcc.org/nomograms.

RESULTS

For all 306 patients, the median overall survival (OS) was 40 months and the median DSS 41 months. Median follow-up for patients alive at last follow-up was 48 months. Lymph node involvement, resection margin status, and tumor differentiation were independent prognostic factors in the derivation cohort (MSKCC). A nomogram with these prognostic factors had a concordance index of 0.73 compared with 0.66 for the AJCC staging system. In the validation cohort (AMC), the concordance index was 0.72, compared with 0.60 for the AJCC staging system. Calibration was good in the derivation cohort; in the validation cohort patients had a better median DSS than predicted by the model.

CONCLUSIONS

The proposed nomogram to predict DSS after curative intent resection of PHC had a better prognostic accuracy than the AJCC staging system. Calibration was suboptimal because DSS differed between the two institutions. The nomogram can inform patients and physicians, guide shared decision making for adjuvant therapy, and stratify patients in future randomized, controlled trials.

Quality of life in RCT of pasireotide to reduce pancreatic complications following resection

393

Background: A recent trial demonstrated that prophylactic pasireotide reduces pancreatic complication (PC) rates (primary objective). In this secondary analysis, we describe quality of life (QoL) in this population before and after resection using a standard instrument (the EORTC QLQ-C30) and the EORTC pancreatic cancer module (PAN26), which was recently developed to address the unmet need for QoL assessment in this subpopulation. Also of interest is assessing the association between PC and QoL and testing the hypothesis that pasireotide improves QoL. Methods: We conducted a randomized, double-blind, placebo-controlled trial of preoperative subcutaneous pasireotide in 300 patients undergoing pancreaticoduodenectomy or distal pancreatectomy. Participants completed the C30 and PAN26 preoperatively and on postoperative days 14 and 60. Scores were compared using paired t-tests. Results: All patients completed at least one questionnaire and 87% of patients completed all three. No major differences in QoL were seen between treatment groups so pooled results were reported. A significant worsening of function at 14 days was detected on all PAN26 and C30 function scales except hepatic and emotional functioning (EF), and all C30 symptom scales (all p<.01). These effects lessened by 60 days, but scores remained significantly worse than baseline with the exception of the sexuality, cognitive functioning, nausea and vomiting, insomnia and constipation scales, which returned to near baseline, and EF, which was significantly better than baseline (p=0.03). PC occurred in 45 patients and was associated with worse body image, dyspnea, financial difficulties and physical, role, emotional and social functioning at 14 and 60 days (all p<.05). Conclusions: During the first 14 days following resection, patients can be expected to have a significant decline in QoL. Many symptoms abate by 60 days, and EF improves. Although pasireotide effectively reduced PC, its effect did not appear to translate to improved QoL based on this sample of 300 patients. While PC was associated with worse QoL, most patients in both pasireotide and placebo groups did not experience PC, which may explain why no significant difference in QoL was observed. Clinical trial information: NCT00994110.

The management of small asymptomatic pancreatic neuroendocrine tumors: A matched case-control study

239

Background: Overdiagnosis and overtreatment has become an evolving challenge for several cancer sub-types. We hypothesized that a substantial portion of incidentally diagnosed small pancreatic neuroendocrine tumors(PanNET) are overtreated as a result of overdiagnosis and that non-operative management may be reasonable for selected patients. Methods: Consecutive patients evaluated for incidentally discovered, sporadic, stage I-II PanNET were analyzed retrospectively. Diagnosis was determined either by pathology or unequivocal imaging characteristics. Patients selected for radiographic surveillance (RS) were matched with patients who underwent resection based on tumor size at initial imaging. Clinicopathological characteristics were compared between the groups. Results: During the study period (2000-2013), RS was recommended for 80 patients, and 79 matched patients underwent resection (resection group). Pathologic diagnosis was obtained in 42 (53%) of the 80 RS patients. Median initial tumor size was similar between the RS vs resection groups (1.2cm (0.8-1.7) vs 1.3 cm (1-1.9), respectively, p=0.4). The resection group was younger and had a longer median follow-up compared to the RS group (58 vs 65 years, p<0.001; 50 vs 29 months, p=0.006; respectively). At the time of last follow-up of the RS group, median tumor size had not changed (1.2cm, p=0.4), no patient had developed metastases, and no patient had experienced radiographic changes in the primary tumor that prompted resection. Within the resection group, low-grade (G1) pathology was recorded in 74 (95%) tumors, one patient had node positive disease, and five developed recurrence (6%). The postoperative complication rate was 36%. No patient in either group died from disease. Death from other causes occurred in 7 out of 159(4%) patients. Conclusions: In this study, no patient who was selected for observation developed metastases or died from disease after a median follow-up of almost 2.5 years.Radiographic surveillance for stable, small, incidentally discovered PanNETs is reasonable in selected patients.

Long-term outcomes following microwave ablation for liver malignancies

BACKGROUND

Microwave ablation has emerged as a promising treatment for liver malignancies, but there are scant long-term follow-up data. This study evaluated long-term outcomes, with a comparison of 915-MHz and 2.4-GHz ablation systems.

METHODS

This was a retrospective review of patients with malignant liver tumours undergoing operative microwave ablation with or without liver resection between 2008 and 2013. Regional or systemic (neo)adjuvant therapy was given selectively. Local recurrence was analysed using competing-risk methods with clustering, and overall survival was determined from Kaplan-Meier curves.

RESULTS

A total of 176 patients with 416 tumours were analysed. Colorectal liver metastases (CRLM) comprised 81.0 per cent of tumours, hepatocellular carcinoma 8.4 per cent, primary biliary cancer 1.7 per cent and non-CRLM 8.9 per cent. Median follow-up was 20.5 months. Local recurrence developed after treatment of 33 tumours (7.9 per cent) in 31 patients (17.6 per cent). Recurrence rates increased with tumour size, and were 1.0, 9.3 and 33 per cent for lesions smaller than 1 cm, 1-3 cm and larger than 3 cm respectively. On univariable analysis, the local recurrence rate was higher after ablation of larger tumours (hazard ratio (HR) 2.05 per cm; P < 0.001), in those with a perivascular (HR 3.71; P = 0.001) or subcapsular (HR 2.71; P = 0.008) location, or biliary or non-CRLM histology (HR 2.47; P = 0.036), and with use of the 2.4-GHz ablation system (HR 3.79; P = 0.001). Tumour size (P < 0.001) and perivascular position (P = 0.045) remained significant independent predictors on multivariable analysis. Regional chemotherapy was associated with decreased local recurrence (HR 0.49; P = 0.049). Overall survival at 4 years was 58.3 per cent for CRLM and 79.4 per cent for other pathology (P = 0.360).

CONCLUSION

Microwave ablation of liver malignancies, either combined or not combined with liver resection, and selective regional and systemic therapy resulted in good long-term survival. Local recurrence rates were low after treatment of tumours smaller than 3 cm in diameter, and those remote from vessels.

Response rates to hepatic arterial infusion (HAI) pump therapy in patients with metastatic colorectal cancer liver metastases (mCRC LM) after progression on all standard chemotherapies

597

Background: To evaluate the overall response rate of HAI with floxuridine (FUDR) in the refractory setting in patients with mCRC LM. Methods: After obtaining an IRB waiver a computerized search was performed for patients with mCRC treated with 5FU, oxaliplatin and irinotecan +/- EGFR and VEGF inhibitor from 2003-2012. Charts were reviewed to ensure patients (pts) had received all standard therapies prior to HAI pump placement. All pts received HAI FUDR and no new systemic targeted or cytotoxic therapies were used with HAI pump. Imaging was re-reviewed for confirmation of progression prior to HAI pump placement and for best response using RECIST 1.1 criteria. Results: 75 pts were identified; of these 23 had radiographic disease progression on all standard chemotherapies (5FU, irinotecan and oxaliplatin) prior to having a pump placed. Of the 23 evaluable pts, the median age was 53 (range 37-75). Six pts had low volume extrahepatic metastases at the time of pump placement. The overall response rate (ORR) was 8/23 (35%); 10/23 (43%) pts had stable disease (SD). The median duration of SD was 4 months (range 1-10). Median follow up, measured from the date of HAI initiation, was 24 months and median overall survival (OS) was 22 months (95% CI 13-16). The median hepatic progression free survival (hPFS) was 4.5 months [CI: 3.8-6.7]. Thirteen pts developed extrahepatic disease progression (including 5 pts with preexisting extrahepatic disease). The median overall PFS was 3.9 months [95% CI 2.24-5.33]. Median number of HAI treatments was 4 cycles (range 1-13). Six out of 23 (26%) pts required a 50-75% dose reduction by the second cycle due to elevated liver function tests and 18/23 (78%) required a dose reduction after the 3rd cycle. No pts required stents or developed long term liver or biliary toxicity. Conclusions: In a cohort of 23 patients with mCRC LM, refractory to all standard agents, treatment with HAI FUDR resulted in an ORR by RECIST 1.1 of 35% and median OS of 22 months. Further studies focusing on locoregional therapy in patients with liver predominant disease are warranted. Studies to molecularly characterize these tumors are ongoing.

Hepatic arterial nodal metastases in pancreatic cancer: is this the node of importance?

BACKGROUND

The hepatic artery lymph node (HALN) is frequently sampled during pancreaticoduodenectomy (PD). Data suggest that survival in the setting of HALN metastases is similar to that of stage IV pancreatic ductal adenocarcinoma (PDAC). The objectives of this study were to describe the prognostic significance of HALN metastases and to assess the predictive performance of HALN compared to peripancreatic lymph node status.

METHODS

Patients undergoing PD for PDAC from January 2000-October 2010 were identified from a prospectively maintained database. Patients were included if during PD the HALN was submitted for pathologic evaluation. Patients were excluded if margins were macroscopically positive, if pathology was found to be consistent with a diagnosis other than PDAC. Overall (OS) and disease-free survival (DFS) were estimated by Kaplan-Meier methods.

RESULTS

Of the 671 patients who underwent PD for PDAC, HALN status was analyzed for 147 patients. HALN was positive in 23 patients (16 %), 38 were peripancreatic lymph node (PPLN) and HALN negative, and 86 were PPLN+/HALN-. Median follow-up for survivors was 10 months. In a multivariable model, lymph node status and tumor differentiation predicted OS and DFS. Hazard of death and relapse/death were highest among the HALN+ patients (hazard ratio [HR] 2.94; p = 0.017 and HR 2.66; p = 0.011, respectively). Kaplan-Meier analysis revealed significant differences in OS (p = 0.017) and DFS (p = 0.013) based on lymph node status.

CONCLUSIONS

OS and DFS are significantly reduced in patients with a positive HALN. Differentiation and lymph node status were predictors of OS and DFS. In the multivariate models, differentiation and lymph node status remain independent predictors of OS and DFS.

Hepatic arterial infusion for unresectable intrahepatic cholangiocarcinoma: An update on survival from two prospective clinical trials

4116

Background: Patients with unresectable intrahepatic cholangiocarcinoma (IHC) experience poor survival. This study summarizes the long-term outcome of two previously reported clinical trials using hepatic arterial infusion (HAI) with floxuridine (FUDR) and dexamethasone (Dex) (with or without bevacizumab (Bev)) in advanced IHC. Methods: Prospectively collected clinicopathologic and survival data were retrospectively reviewed. Disease response was based on RECIST. All patients underwent pre-treatment dynamic contrast enhanced MRI (DCE-MRI), and tumor perfusion data were correlated with outcome. Results: Forty-four patients were analyzed (FUDR=26, FUDR/Bev=18). At a median follow-up of 30 months, 41 patients had died of disease and 3 were alive. Partial response was observed in 48% of patients, and another 50% had stable disease. Three patients underwent resection after HAI and 84% received additional HAI after removal from the study. Median survival was similar in both trials (FUDR=29 months vs. FUDR/Bev=28.5 months p=0.96). Ten patients (23%) survived ≥3 years including 5 (11%) ≥ 5 years. Tumor perfusion, as measured on pre-treatment DCE-MRI (area under the gadolinium concentration curve (AUC180)), was significantly higher in ≥3-year survivors, and was the only factor that distinguished this group from <3-year survivors (mean AUC180 48.9mM.s vs 32.3mM.s, respectively; p=0.003). Time to liver progression was longer in ≥3-year survivors (19.8 months vs 11.2 months, respectively; p=0.02). Conclusions: HAI chemotherapy can result in prolonged survival in unresectable IHC. Pre-treatment DCE-MRI may predict response and survival. [Table: see text]

Association of KRAS mutation with worse recurrence-free survival and site of metastatic progression after resection of hepatic colorectal metastases

3609

Background: There are conflicting results regarding the influence of KRAS mutation status and outcome in patients (pts) with colorectal cancer. A recent report suggested worse outcome in KRAS mutated (MUT) pts who underwent resection of hepatic metastases (Karagkounis et al, ASCO 2012). Methods: Recurrence patterns and survival were evaluated in 169 patients who had undergone resection of liver metastases, then received adjuvant hepatic arterial infusion and systemic chemotherapy, and for whom KRAS data were available. Kaplan-Meier methods were used to estimate recurrence free survival (RFS) and overall survival (OS). Log-rank test was used to determine whether survival functions differed by KRAS mutation status. Cumulative incidence function was used to estimate the probability of time from adjuvant therapy to bone, brain, lung and liver metastases separately. Mutations in KRAS (codons 12, 13) were detected using the iPLEX assay (Sequenom, Inc). Results: Median follow-up for the entire cohort was 38.8 months. 118 were KRAS wildtype (WT), and 51 were KRAS MUT (45 G12, 5 G13, 1 K117N). The 3 year RFS was 48% [95%CI: 37-58%] for KRAS WT pts and 30% [15-44%] for MUT pts (p<0.01). OS at 3 years was 96% [88-98%] for KRAS WT and 80% [61-90%] for MUT pts (p=0.08). Cumulative incidence of developing bone, brain, lung, and liver metastases by 2 years is presented in Table 1. The cumulative incidence of metastases to bone at 2 years was 0% and 13.7% in KRAS WT versus MUT pts (p<0.01), to brain 0% versus 4.6% for KRAS WT versus MUT (p=0.05), and to lung 27% versus 47.5% in KRAS WT versus MUT pts (p<0.01). Conclusions: In pts who have had liver resection followed by adjuvant therapy, those with KRAS MUT have a worse RFS and seemingly worse OS than those who are KRAS WT. Also, patients with KRAS MUT appear more likely to develop bone, brain, and lung metastases. Further investigation of a larger number of patients is warranted. [Table: see text]

Survival after resection plus intra-operative radiofrequency ablation (IRFA) to treat colorectal liver metastases (CLM): Results of an international collaborative study

3558

Background: Adding IRFA to parenchymal resection to treat CLM is gaining increasing acceptance in specialized HPB teams treating complex, bilobar disease. Objectives were to confirm the promising results of the prospective CLOCC and ARF2003 trials on a larger international scale. Methods: Four centers combined their clinical databases regarding IRFA for CLM. Demographics, treatments, CLM characteristics, complications (Clavien-Dindo), local recurrence, and survivals (liver progression-free, LPFS, relapse-free, RFS and overall, OS) were analyzed. Results: 280 patients (38% female, median age 61y) received resection plus IRFA over 2001-2011. 205 had synchronous CLM (73%) and 247 bilateral (88%). 227 patients received pre-operative chemotherapy (173 one line, 37 two lines, 10 three lines, 7 missing); 189 received post-operative chemotherapy (103 one line, 46 two lines, 40 three lines). Median number of tumors resected was 2 (range 1-19) and ablated 2 (1-12). Median size (mm) of largest CLM ablated per patient was 8.5(0.1-50). 96 patients experienced complications: 29 G1, 19 G2, 35 G3, 10 G4, and 3 deaths. 48 patients had local recurrence of ablated CLM. 155 patients developed new CLM, 165 extra-hepatic metastases, and 119 patients died during follow-up. One-year, 3-year and median (months) RFS, LPFS and OS were respectively: RFS 41%(95CI35-47), 14%(95CI9-19), 9m (95CI8-11); LPFS 53%(95CI47-59), 31%(95CI25-37), 15m (95CI11-19); OS 90%(95CI85-93), 58%(95CI51-65), 40m (95CI37-50). Median follow-up was 38m (95CI34-49). Conclusions: In this difficult-to-treat group, survival results were good and comparable with rates reported after resection only. IRFA complements resection, enabling to treat more patients, and offers the advantage of sparing healthy parenchyma.

Perioperative complications influence recurrence and survival after resection of hepatic colorectal metastases

BACKGROUND

Perioperative outcomes, such as blood loss, transfusions, and morbidity, have been linked to cancer-specific survival, but this is largely unsupported by prospective data.

METHODS

Patients from a previous, randomized trial that evaluated acute normovolemic hemodilution during major hepatectomy (≥3 segments) were reevaluated and those with metastatic colorectal cancer (n = 90) were selected for analysis. Survival data were obtained from the medical record. Disease extent was measured using a clinical-risk score (CRS). Log-rank test and Cox proportional hazard model were used to evaluate recurrence-free survival (RFS) and overall survival (OS).

RESULTS

Median follow-up was 71 months. The CRS was ≥3 in 45 % of patients; 59 % had extrahepatic procedures. Morbidity and mortality were 33 and 2 %, respectively. Postoperative chemotherapy was given to 87 % of patients (78/90) starting at a median of 6 weeks. RFS and OS were 29 and 60 months, respectively. Postoperative morbidity significantly reduced RFS (23 vs. 69 months; P < 0.001) and OS (28 vs. 74 months; P < 0.001) on uni- and multi-variate analysis; positive resection margins and high CRS also were significant factors. Delayed initiation of postoperative chemotherapy (≥8 weeks) was common in patients with complications (37 vs. 12 %; P = 0.01).

CONCLUSIONS

In this selected cohort of patients from a previous RCT, perioperative morbidity was strongly (and independently) associated with cancer-specific outcome. It also was associated with delayed initiation of postoperative chemotherapy, the impact of which on survival is unclear.

Influence of KRAS mutation on recurrence patterns in patients undergoing hepatic resection of colorectal metastases

398

Background: There have been conflicting results about whether KRAS mutation influences outcome in patients (pts) with colorectal cancer. In pts who underwent liver resection, Karagkounis reported a worse recurrence and survival in KRAS mutated (MUT) patients (ASCO 2012, abs 3616). Methods: In 105 pts who underwent liver resection and received adjuvant (adj) hepatic arterial infusion and systemic chemotherapy and in whom KRAS data was available, we evaluated recurrence patterns and survival. Correlation between KRAS and clinical factors such as prior chemotherapy, post operative CEA, clinical risk score, and stage at diagnosis was evaluated using Fisher’s exact test and the Wilcoxon rank sum test. Kaplan-Meier methods were used to estimate median overall recurrence free survival (RFS) and overall survival (OS) at 4 years. Log-rank test was used to determine whether survival functions differed by KRAS mutation status. Cumulative incidence function was used to estimate the probability of time from adj therapy to bone, brain, lung and liver metastases separately. Results: Of 105 patients, 76 were KRAS wildtype (WT), and 29 were KRAS MUT (26-G12 and 3-G13). The median RFS was 26 months for KRAS WT pts and 15 months for KRAS MUT pts (p=0.08). OS at 4 years was 88% [95% CI: 78%-94%] for KRAS WT and 78% [95% CI: 57%-90%] for KRAS MUT pts (p= 0.15). Cumulative incidence of developing bone, brain, lung, and liver metastases by 2 years is presented in the Table. The cumulative incidence of bone and brain metastases at 2 years was 0% and 0% in KRAS WT pts versus 16.4% [95% CI: 1.1%-31.7%] and 4.7% [95% CI: 0%-14.1%] in KRAS MUT pts (Table). There was no association between clinical factors and KRAS status. Conclusions: KRAS MUT pts appeared to have worse OS and RFS, although we were unable to show a significant difference between KRAS WT and MUT for OS and RFS. In addition, cumulative incidence of bone and brain metastases at 2 years appeared to be higher for KRAS MUT pts as compared to WT pts. Results are based on small sample size and further investigation is needed. [Table: see text]

Malignant progression in IPMN: a cohort analysis of patients initially selected for resection or observation

BACKGROUND

Intraductal papillary mucinous neoplasms (IPMN) may represent a field defect of pancreatic ductal instability. The relative risk of carcinoma in regions remote from the radiographically identified cyst remains poorly defined. This study describes the natural history of IPMN in patients initially selected for resection or surveillance.

METHODS

Patients with IPMN submitted to resection or radiographic surveillance were identified from a prospectively maintained database. Comparisons were made between these two groups.

RESULTS

From 1995 to 2010, a total of 356 of 1,425 patients evaluated for pancreatic cysts fulfilled inclusion criteria. Median follow-up for the entire cohort was 36 months. Initial resection was selected for 186 patients (52 %); 114 had noninvasive lesions and 72 had invasive disease. A total of 170 patients underwent initial nonoperative management. Median follow-up for this surveillance group was 40 months. Ninety-seven patients (57 % of those under surveillance) ultimately underwent resection, with noninvasive disease in 79 patients and invasive disease in 18. Five of the 18 (28 %) invasive lesions developed in a region remote from the monitored lesion. Ninety invasive carcinomas were identified in the entire population (25 %), ten of which developed the invasive lesion separate from the index cyst, representing 11 % with invasive disease.

CONCLUSIONS

Invasive disease was identified in 39 % of patients with IPMN selected for initial resection and 11 % of patients selected for initial surveillance. Ten patients developed carcinoma in a region separate from the radiographically identified IPMN, representing 2.8 % of the study population. Diagnostic, operative, and surveillance strategies for IPMN should consider risk not only to the index cyst but also to the entire gland.

Effect of perioperative complications on recurrence and survival after resection of hepatic colorectal metastases: Analysis of data from a randomized controlled trial

3620

Background: Perioperative outcomes such as blood loss, transfusions and morbidity have been associated with cancer specific survival, but this is unsupported by prospective data. Methods: Patients were derived from a previously reported prospective randomized trial of acute normovolemic hemodilution (ANH) versus standard management during major hepatectomy. Patients with metastatic colorectal cancer (mCRC) were selected for analysis. Chemotherapy and survival data were obtained from the medical record. Disease extent was measured using a clinical risk scoring (CRS) system. Log-rank test and Cox proportional hazard model were used to evaluate recurrence free (RFS) and overall survival (OS). Results: This trial enrolled 130 patients, 90 of whom had mCRC. Median follow up was 54 m; median age was 53 yr; 56% were men. The CRS was ≥3 in 55% of patients, and 57% had additional extrahepatic procedures. Morbidity and mortality were 33% and 3%. Chemotherapy was split before and after surgery in 70%; 10% received only preoperative and 19% only postoperative treatment, which was initiated after a median time of 6 wks (IQR = 4-7). RFS and OS were 29 and 59 m. On multivariate analysis, (+) resection margin, high CRS and perioperative complications predicted shorter RFS and OS. Delayed initiation of postoperative chemotherapy (≥ 8 wks) while most common in patients with complications (37 vs 12%; p: 0.01), was not a significant predictor of shorter RFS or OS. Randomization (ANH vs standard) was also not significant in this regard. (See Table.) Conclusions: In this prospective randomized cohort, perioperative morbidity was a highly significant independent predictor of cancer specific outcome, associated with but not entirely explained by delayed initiation of chemotherapy. [Table: see text]

Is conversion to resection possible with hepatic arterial infusion (HAI) and systemic (SYS) even in previously treated patients (pts) with unresectable colorectal liver metastases (UnCLM)?

3577

Background: Previously, we showed thatHAI with FUDR + dexamethasone (Dex) plus SYS produced a 47 % resectability rate in a retrospective study of 49 pts with UnCLM. Methods: Prospectively evaluated UnCLM pts in a new protocol were combined with the above protocol (n=105 pts) and all were treated with HAI FUDR/Dex + Sys. Unresectability was defined as diffuse bilateral metastases, involvement of all hepatic/portal veins, and/or inability to preserve remaining liver with adequate perfusion. Factors associated with conversion were identified using a multivariate logistic regression model. Overall survival (OS) and progression free survival (PFS) were calculated from pump placement by the Kaplan-Meier method. Resectability was a time-dependent covariate in a Cox regression model. Results: 61 of the 105 pts had prior SYS (56 %with prior Oxali) and 45 (74%) were progressing at the time of pump placement. In previously treated pts, 44% underwent resection, with a median OS of 45 mos. Of 44 chemo-naïve pts, 57% underwent hepatectomy, with a median OS of 68 mos. The following were significantly associated with resection conversion: lesion number [p=0.02], baseline CEA [p=0.04], females [p=0.03] and clinical risk score (CRS) [p=0.05]. In multivariate analysis, gender and CRS remained predictive of resectability. Surgery greatly reduced the hazard of death by 67% [HR: 0.33, 95%CI: 0.17-0.61, p=0.0004], after adjusting for several risk factors (Table). Median PFS was 12 mos for all pts. Conclusions: Even in previously treated pts,HAI + SYS is an approach to convert UnCLM to resection. Gender and CRS are associated with conversion to resectability. [Table: see text]

Use of detailed pathologic characteristics of the primary colorectal tumor to predict outcome after hepatectomy for metastases

559

Background: Outcome after hepatic resection for colorectal liver metastases (CRLM) is heterogeneous and accurate predictors of survival are lacking. The aim of our study was to analyze the prognostic relevance of pathologic details of the colorectal primary tumor in patients undergoing hepatic resection of CRLM.

Methods: Retrospective review of a prospective database identified patients who underwent potentially curative hepatic resection of CRLM. Clinicopathological variables were investigated and their association with outcome was analyzed.

Results: From 1997-2007, 1004 patients underwent hepatic resection for CRLM. The median follow-up for survivors was 59 months with a 5-year predicted survival of 47%. Ninety-two percent of patients received perioperative chemotherapy and 34% received adjuvant hepatic artery infusion chemotherapy. Univariate analysis identified 10 factors associated with poor survival. Three of these related to the pathology of the primary tumor: lymphovascular invasion (LVI, p<0.0001), perineural invasion (p=0.005), and degree of regional lymph node involvement (N0 vs N1 vs N2, p<0.0001). Multivariate analysis identified 7 factors associated with poor survival. Two of these related to the pathology of the primary tumor: LVI (HR 1.3, 95% CI 1.06-1.64, p=0.01) and degree of regional lymph node involvement where an increase in the number of metastatic regional lymph nodes from N1 (HR 1.3, 95% CI 1.04-1.69, p=0.02) to N2 (HR 1.7, 95% CI 1.27-2.21, p <0.0005) was associated with a reduced survival. LVI positive patients had a median survival of 48 months compared to 69 months for LVI negative patients (p<0.0001); moreover, patients who were LVI positive with N2 nodal disease had a reduced survival to 40 months compared to 74 months for patients who were LVI negative with no nodal disease (p<0.0001).

Conclusions: Resection of CLRM is associated with long-term survival. Pathologic details of the primary colorectal tumor, particularly LVI and the degree of lymph node involvement, are strong predictors of survival. Future biomarker studies should consider utilizing factors related to the primary colorectal tumor.

Malignant progression in intraductal papillary mucinous neoplasms of the pancreas: Results of 157 patients selected for radiographic surveillance

152

Background: Natural history studies suggest that intraductal papillary mucinous neoplasm (IPMN) is a field defect of pancreatic ductal instability. The risk of malignancy is well known, but it is unclear if the radiographic abnormality is the predominant site at which this progression occurs. This study analyzes the prevalence and site of pancreatic ductal adenocarcinoma (PDAC) progression in patients initially selected for radiographic surveillance.

Methods: Patients evaluated for pancreatic cystic lesions between 1995-2010 were reviewed. Patients were included if they were followed for > 6 months for a cystic lesion with either a documented cyst fluid CEA >/= 200 ng/mL or pathologic confirmation of an IPMN.

Results: Of the 157 patients initially selected for surveillance, 97 (62%) eventually underwent resection. The median length of surveillance prior to operation was 15 mo (range: 6-193 mo). Pathologically confirmed carcinoma in situ (high grade dysplasia, n=22) or invasive carcinoma (n=18) was identified in 40 patients. Of the 18 patients who were found to have invasive carcinoma, 10 had main duct IPMN (56%), 5 had branch duct (28%), and 3 had combined (17%). Four of the 18 patients who developed invasive cancer during surveillance (22% of those resected for carcinoma and 3% of those followed) developed PDAC in a region of the gland distinct from the radiographically identified lesion for which surveillance was recommended. Invasive carcinoma was diagnosed a median of 24 mo after the original IPMN diagnosis. During follow-up (median = 8 months from the cancer diagnosis), 1 patient died of disease, 2 are cancer-free, and 1 patient is alive with recurrence.

Conclusions: In this study, 11% of patients with IPMN initially selected for surveillance developed invasive disease; 22% of those tumors arose in an area of the gland distinct from the initially identified lesion. Diagnostic and operative strategies for IPMN should consider the cancer risk in the entire gland.

[Table: see text]

Conversion to complete surgical treatment using hepatic artery infusional chemotherapy in patients with unresectable liver metastases from colorectal cancer

472

Background: Hepatic artery infusional (HAI) chemotherapy effectively treats colorectal liver metastases (CRLM). This study analyzes the combination of HAI and systemic chemotherapy for treating unresectable CRLM, focusing on the conversion to complete resection and/or ablation (R/A) and long term outcomes.

Methods: All patients with initially unresectable CRLM treated with HAI and systemic chemotherapy from 2000-2009 were included. Patients who responded sufficiently to undergo complete R/A were compared to those who did not convert.

Results: 373 patients were included. 296 (79 %) were previously treated and 77 (21 %) were chemo naïve. 115 (31 %) were on protocol and 258 (69 %) were not on protocol. 93 patients (25%) subsequently underwent complete R/A (conversion group). Of the 115 protocol patients, 47 (41 %) underwent complete R/A. The percentage of patients submitted to complete R/A increased from 16% during 2000-2003 to 30% during 2004-2009. 43% of patients who were chemotherapy-naïve prior to HAI therapy eventually underwent complete R/A, compared to 15% who were initially treated with systemic therapy (p<0.001). Factors associated with conversion on multivariate analysis were more recent treatment (2004-2009), no prior chemotherapy, and a clinical risk score < 3. Overall survival was greater in the conversion group, with a median and predicted 5-year survival from the time of HAI pump placement of 54 months and 49%, compared to 19 months and 6%, respectively (p<0.001).

Conclusions: One-quarter of patients with unresectable CRLM responded sufficiently to undergo complete R/A following HAI plus systemic chemotherapy. Conversion was more likely in patients who were chemotherapy-naïve (43 %), had a clinical risk score < 3, and were treated in the more recent era. Survival in the conversion group was significantly better.

Significance of CEACAM6 expression in biliary tract carcinoma

207

Background: The entire biliary tree is at risk for malignant change, but little is known about differences in molecular pathogenesis with respect to anatomic site. CEACAM6 is a membrane protein involved in cell adhesion and signaling that is overexpressed in pancreatic adenocarcinoma and associated with poor prognosis. This study examines CEACAM6 expression in the entire spectrum of biliary carcinomas and its relationship to outcome.

Methods: Tissue microarrays containing triplicate cores of paraffin-embedded surgical specimens from patients with bile duct carcinoma (hilar, intrahepatic, distal) and control tissue were stained for CEACAM6 by immunohistochemistry. Clinical, pathologic and survival data were analyzed and correlated with CEACAM6 expression. Survival was estimated using the Kaplan-Meier method and compared with log rank test.

Results: One hundred twenty cases of bile duct carcinomas from 1992-2007 were assembled in the tissue microarrays. Strong CEACAM6 signal was present in 30/60 (50%) of hilar tumors, 7/45 (16%) of intrahepatic tumors, 7/15 (47%) of distal tumors, and none of the control tissues. Overall median survival and follow-up were 36.4 months and 98.3 months, respectively. CEACAM6 staining did not correlate with sex, grade, positive lymph nodes, vascular invasion or metastases but was associated with age > 65 (p<0.05) and higher T stage (p<0.05). After R0 resection, CEACAM6 expression was associated with disease-specific survival (DSS) only in the subset of patients with intrahepatic cholangiocarcinoma (median DSS 78 months for negative and 16 months for positive, p<0.002). Vascular invasion was the sole independent predictor of survival on multivariate proportional hazards regression (HR=1.742 [1.048-2.895 95%CI], p<0.03) in the entire cohort.

Conclusions: CEACAM6 may serve as a marker of poor outcome in patients with intrahepatic cholangiocarcinoma and should be further evaluated as a means of selecting patients for adjuvant therapy after resection.