Hepatic metastasis from colorectal cancer

Cytotoxic agents (see Chapter 98 )

Since the 1950s, 5-FU has been the cornerstone of treatment for metastatic CRC. Leucovorin (LV), also known as folinic acid, enhances the activity of 5-FU by stabilizing binding of 5-FU to its target, thymidylate synthase. Irinotecan, a topoisomerase inhibitor, was approved by the United States Food and Drug Administration (FDA) in 2000 for first-line treatment of metastatic CRC, based on two randomized controlled trials (RCTs) showing improved response and OS rates with the addition of irinotecan to 5-FU/LV. ^, Oxaliplatin is a platinum derivative that exerts synergistic cytotoxicity with 5-FU and received FDA approval in 2004 for first-line treatment of metastatic CRC. ^,

Regimens with a backbone of infusional 5-FU/LV, combined with oxaliplatin (FOLFOX) or irinotecan (FOLFIRI), have demonstrated objective response rates greater than 50% in the first-line treatment of metastatic CRC. Four RCTs have demonstrated that FOLFOX and FOLFIRI have similar clinical efficacy, and the choice of regimen can be tailored to the individual patient and toxicity profiles. Capecitabine is an oral fluoropyrimidine that is converted to 5-FU in tumor tissues that can be combined with oxaliplatin (XELOX) or irinotecan (XELIRI).

The combination of 5-FU, LV, oxaliplatin, and irinotecan (FOLFOXIRI) for unresectable metastatic CRC was compared with FOLFIRI in a phase III RCT by the Gruppo Oncologico Nord Ovest. FOLFOXIRI was associated with significantly higher response rates, progression-free survival (PFS), and OS. However, toxicity was greater in the FOLFOXIRI arm, with statistically significant greater incidence of grade 2 to 3 neuropathy and grade 3 to 4 neutropenia.

Biologic therapies

Biologic agents in combination with first-line chemotherapy have shown promising results. Bevacizumab, a monoclonal antibody against vascular endothelial growth factor, is associated with improved survival when added to irinotecan or oxaliplatin-based regimens in the first-line metastatic setting. ^, Cetuximab is a monoclonal antibody directed against epidermal growth factor receptor (EGFR). Mutations in KRAS , which lies downstream in the EGFR signaling pathway, are associated with resistance to cetuximab. For KRAS wild-type tumors, the addition of cetuximab to FOLFIRI or FOLFOX is associated with improved response rates and PFS. ^, Panitumumab is another monoclonal antibody directed against EGFR, which has similar efficacy and toxicity as cetuximab.

Metastatic CRC patients with BRAF V600E mutations have particularly poor outcomes, with median OS of only 4 to 6 months after failure of initial treatment. In CRC, monotherapy with BRAF inhibitors is ineffective because of pathway reactivation through EGFR signaling. The recently published BEACON trial showed improved OS and response rates with triplet therapy of encorafenib, binimetinib, and cetuximab, over the control group of cetuximab and irinotecan. Combination therapy with encorafenib, a BRAF inhibitor, anti-EGFR therapy (cetuximab), and binimetinib, a selective inhibitor of mitogen-activated protein kinase (MAPK) kinase, overcomes the limitations of BRAF inhibition alone.

Perioperative chemotherapy for resectable disease

After potentially curative resection of CLM, two-thirds of patients suffer disease recurrence, and half of these recurrences occur in the liver. In efforts to improve relapse rates, perioperative chemotherapy was compared with surgery alone in the European Organisation for Research and Treatment of Cancer (EORTC) intergroup trial 40983 (EPOC). Patients with resectable CLM, up to 4 in number, were randomized to perioperative FOLFOX or hepatectomy alone. Among 342 eligible patients, median PFS was 20.9 months in the perioperative chemotherapy arm, compared with 12.5 months in the surgery alone arm ( P = .035). OS rates were similar between the two arms. More recently, the New EPOC study showed a detrimental effect on survival with the addition of cetuximab to perioperative chemotherapy for patients with KRAS exon 2 wild-type, resectable CLM.

In the adjuvant setting, an RCT of adjuvant 5-FU/LV versus surgery alone showed improved 5-year disease-free survival (DFS) with adjuvant treatment (33.5% vs. 26.7%, P = .028) but no statistically significant benefit in OS. This trial was suspended because of slow accrual and enrolled 173 patients from 47 hospitals in a 10-year time period. In a subsequent randomized study, FOLFIRI compared with 5-FU/LV after CLM resection had no significant impact on DFS or OS. These results are congruent with the lack of efficacy of FOLFIRI for adjuvant treatment of stage III CRC. ^,

Conversion therapy for borderline resectable disease

A study published in 2004 of over 1000 CLM patients showed that 12.5% of patients with initially unresectable CLM achieved sufficient downsizing with systemic therapy to undergo hepatic resection. Among these patients with initially unresectable disease, 5-year OS after hepatectomy was 33%. Most of the patients in this study received oxaliplatin-based chemotherapy. FOLFOXIRI has been shown in RCTs to improve resectability rates over FOLFOX or FOLFIRI. ^, However, hepatotoxicity with FOLFOXIRI can be significant, as described later in this chapter.

Biologic agents have also been shown to increase resectability rates. In a recent RCT of 241 patients with unresectable, RAS -mutated CLM, the addition of bevacizumab to FOLFOX increased objective response rates and conversion to resectability, with 22.3% of patients undergoing R0 resection after FOLFOX plus bevacizumab, compared with 5.8% of patients after FOLFOX alone ( P < .01). Similarly, cetuximab added to FOLFOX or FOLFIRI for RAS wild-type CLM increased resectability rates in retrospective and prospective studies. ^,

Hepatic arterial infusion chemotherapy (see Chapter 97 )

The rationale for infusion of chemotherapy directly into the hepatic artery is the liver’s dual blood supply, with metastases perfused predominantly via the hepatic artery, in contrast to portal blood supply to normal liver. Thus hepatic arterial infusion (HAI) delivers chemotherapy preferentially to tumor over normal liver parenchyma. Floxuridine (FUDR), a metabolite of 5-FU, has been extensively studied for HAI because of its high first-pass extraction in the liver, thereby minimizing systemic exposure.

In 1999 an RCT of adjuvant 5-FU plus HAI-FUDR compared with 5-FU alone after CLM resection demonstrated significantly higher 2-year OS in the HAI-FUDR group (86% vs. 72%, P = .03). Two-year hepatic recurrence-free survival (RFS) was also significantly higher with HAI-FUDR (90% vs. 60%, P < .001). More recently, a phase II single-arm study of HAI-FUDR plus systemic therapy for 64 patients with initially unresectable CLM reported response and conversion to resectability rates of 73% and 52%, respectively.

Disadvantages of HAI chemotherapy include biliary toxicity, which can be mitigated with the concurrent infusion of dexamethasone and avoidance of systemic administration of bevacizumab. Hepatic toxicity with HAI oxaliplatin reportedly occurs more frequently than with systemic oxaliplatin. In addition, multidisciplinary expertise in surgery, nuclear medicine, interventional radiology, and gastroenterology is required for placement and management of HAI pumps.

Chemotherapy-associated hepatotoxicity

A study published in 2006 from The University of Texas MD Anderson Cancer Center (MDACC) demonstrated associations between oxaliplatin and sinusoidal injury and between irinotecan and steatohepatitis ( Fig. 90.2 ). Oxaliplatin-related sinusoidal injury can progress to nodular regenerative hyperplasia and clinically significant portal hypertension. Splenomegaly and thrombocytopenia are surrogates for oxaliplatin-induced sinusoidal injury. In the MDACC study, steatohepatitis, but not sinusoidal injury, was associated with higher mortality after CLM resection (14.7% vs. 1.6%, P = .001; see Chapter 98 ).

FOLFOXIRI is associated with high response rates but correspondingly high rates of liver toxicity. In a pooled analysis of 37 patients who underwent R0 hepatectomy after FOLFOXIRI for initially unresectable disease, sinusoidal dilation and steatosis were identified in 100% and 76% of resected specimens, respectively.

Strategies to mitigate the negative effects of chemotherapy-associated hepatotoxicity include limiting preoperative chemotherapy to up to six cycles, sufficient time interval between chemotherapy and liver resection, and the use of bevacizumab with oxaliplatin to protect from sinusoidal injury. In the EPOC trial, patients who received six cycles of preoperative FOLFOX had higher rates of reversible postoperative complications than those who underwent surgery alone (25% vs. 16%, P = .0401). In a study from Memorial Sloan-Kettering Cancer Center (MSKCC), maximum radiologic response in CLM was observed after 2 to 4 months of chemotherapy; if continued beyond 4 months, there was little gain in therapeutic benefit. Twelve or more cycles of preoperative chemotherapy are associated with significantly higher rates of severe sinusoidal injury, postoperative major morbidity, and mortality.

A longer interval between chemotherapy and CLM resection is associated with lower morbidity but should be balanced with the risk of disease progression during the treatment-free interval. The risk of postoperative complications is reportedly twice as high among patients with interval between chemotherapy and CLM resection of up to 4 weeks, compared with 5 to 8 weeks. Thus an interval of 5 weeks is recommended between chemotherapy and surgery. Despite its antiangiogenic effects, bevacizumab can also be administered 5 weeks before hepatectomy without an increase in postoperative complications.

Treatment sequencing

For patients with resectable and initially unresectable CLM, treatment sequencing integrates short courses of chemotherapy, evaluation of response, tumor genetics, and one- or two-stage hepatectomy (TSH) with portal vein embolization (PVE) when indicated ( Fig. 90.3 ). The choice of first-line therapy with FOLFOX or FOLFIRI depends on toxicity profiles, particularly hepatic sinusoidal injury and peripheral neuropathy with oxaliplatin and steatohepatitis and gastrointestinal toxicity with irinotecan. Patients with baseline splenomegaly ar e at risk for severe sinusoidal injury with oxaliplatin, whereas patients with hepatic steatosis, obesity, and diabetes are at risk for developing steatohepatitis with irinotecan. Given the improved pathologic response with bevacizumab and its protective effect on oxaliplatin-associated sinusoidal injury, FOLFOX plus bevacizumab is a commonly used first-line regimen. ^,

Criteria for resectability

Preoperative evaluation of patients with CLM begins with assessment of their physical fitness to undergo hepatectomy, including chronic comorbidities, acute infectious or thrombotic processes, underlying liver impairment, and performance status. In particular, given the cardiovascular demands of low central venous pressure anesthesia and possible portal pedicle clamping, any history of cardiac or pulmonary disease must be investigated because these patients are at significant risk for intraoperative and postoperative complications (see Chapters 25 , 26 , and 28 ). Age alone does not constitute a contraindication in the absence of other medical factors.

Evaluation of oncologic resectability involves complete radiologic staging, biomarkers and other prognostic factors, response to systemic therapy, and endoscopic evaluation if the primary tumor is intact. The goals of this assessment are to determine that the primary tumor site is either completely treated or amenable to simultaneous or future resection and to quantify the number and location of extrahepatic disease (EHD).

Technical resectability of intrahepatic disease is defined as the ability to achieve a margin-negative resection while preserving adequate biliary drainage and vascular inflow and outflow, and sparing at least two contiguous hepatic segments with future liver remnant (FLR) volume of more than 20% with normal liver, more than 30% after extensive chemotherapy, and more than 40% with cirrhosis.

Preoperative imaging (see Chapter 15 )

High-quality, contrast-enhanced cross-sectional imaging is essential to identify liver metastases and their relationship to major vessels and bile ducts and to evaluate the quality of nontumoral liver and anticipated FLR volume. The quality of baseline imaging is critically important before systemic therapy, which can lead to disappearing metastases and induce changes in the liver parenchyma that reduce the sensitivity of radiologic exams.