Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Management of Secondary Hepatic Neoplasms
Secondary hepatic neoplasms refer to a heterogeneous collection of tumors that metastasize to the liver. By definition, these cancers develop from other organ sites but share a common metastatic pathway. Tumors that hematogenously disseminate to the liver include carcinomas (e.g., colorectal, pancreatic, gastric, breast, lung), neuroendocrine cancers, and certain types of retroperitoneal sarcomas and gastrointestinal stromal tumors. Systemic chemotherapy may be associated with improved survival compared with untreated patients, but it rarely results in cure. The enhanced efficacy of systemic chemotherapeutic regimens has increased tumor response rates and improved the progression-free and overall survival of patients with these malignancies. The ability to effectively control systemic disease with chemotherapy and reduce either the size of large or the number of diffuse hepatic metastases has expanded the pool of patients eligible to receive curative surgical therapy. Advances in the perioperative management of patients undergoing liver surgery undoubtedly contribute to these improved outcomes. Most importantly, our enhanced collective understanding of the molecular and biologic behavior of these tumors facilitates the appropriate use, combination, and sequencing of cancer-directed treatments. Although much of our knowledge has been drawn from retrospective analyses, several prospective, randomized trials provide an evidence-based rationale for therapy.
As the primary drainage basin of the portal circulatory system, the liver is the most common site of gastrointestinal tract metastases. Colorectal cancer (CRC) is the second leading cause of cancer-specific mortality in the United States and the most frequent tumor type classified as secondary hepatic neoplasm. Up to half of patients with CRC develop hepatic metastases during the course of their disease. Combination treatment with systemic chemotherapy and complete hepatic metastasectomy has been associated with long-term (5-year) survival rates that may approach 50%. A smaller percentage of patients are completely cured of their disease. In neuroendocrine carcinomas with spread to the hepatic parenchyma, complete surgical resection of isolated metastases has been associated with improved survival. Although hepatectomy and locoregional ablative therapies have been used for noncolorectal, nonneuroendocrine (NCNN) metastases, the biology of these tumors is variable, and surgical resection should be reserved for patients with excellent performance status and adequate control of the primary lesion and in select clinical scenarios in which the disease-free interval can be measured in years.
Generally accepted criteria for hepatic metastasectomy include (1) acceptable patient performance status to tolerate the necessary hepatic resection; (2) the liver as the only or predominant site of metastatic disease; (3) the primary tumor (and all other sites of disease) must be completely resectable and not progressing on the most effective systemic chemotherapy regimen; (4) favorable tumor biology, such that rapid progression or widespread micrometastatic disease is unlikely; (5) resection of metastasis will result in either the possibility of a long-term disease-free state or cure. Factors that influence the extent of hepatic resection are the number, size, and location of hepatic lesions, baseline hepatic function, and size of the anticipated postresection liver remnant. Traditionally, the presence of bilobar hepatic metastases and extrahepatic disease (EHD) (in distant nodal basins or lungs) had been considered absolute contraindications to complete hepatic metastasectomy. The combination of effective neoadjuvant chemotherapy, increasingly effective local ablative therapies, and aggressive surgical resection has broadened the indications for surgical resection. The ability to achieve an R0 (negative gross and microscopic margins) resection appears to be a significant factor that is associated with improved disease-free survival rates. Although hepatic recurrence rates have been shown to be higher when the microscopic margins are positive for residual disease, several retrospective reports suggest that this may not influence long-term survival. Positive margins likely reflect poor tumor biology and increased risk of hepatic recurrence. Anatomic resections, along the vascular inflow and outflow structures and biliary ducts, are indicated for metastatic disease in which subanatomic or “wedge” resections would lead to a higher probability of positive margins, major blood loss, or a bile leak. The unique capability of the liver parenchyma to regenerate after hepatectomy allows the surgeon to resect up to 80% of the original volume if baseline hepatic function is normal. The oncologic principles regarding CRC, neuroendocrine, and other hepatic metastasis will be discussed here, as a separate chapter will address the surgical and technical details of liver resection more thoroughly.
Colorectal Cancer Metastases
The liver is the most common site of distant metastatic disease in CRC patients. Approximately half of patients with CRC will develop metastases during their course of disease, and up to 25% will have liver metastases at the time of presentation. The historical results of resection, before the development of current-day chemotherapeutic regimens, typically yielded 5-year survival rates between 20% and 40%. Selected patients undergoing modern chemotherapeutic regimens in combination with complete metastasectomy can achieve durable 5-year survival rates exceeding 50% and as high as 70% in those with solitary liver metastasis. Approximately 17% to 25% of these patients are cured of their disease. These results have established surgery as the standard of care for patients with resectable liver metastasis. These improved survival trends are likely related to improved patient selection with favorable tumor biology, dramatic improvement in the systemic chemotherapeutic regimens, and the concomitant development of new, targeted therapeutics.
The historical selection criteria used to identify stage IV patients who might benefit from metastasectomy have focused on disease burden: number and size of metastatic lesions, bilateral hepatic disease, lymph node–positive primary tumors, and high carcinoembryonic antigen (CEA) levels. The Fong Clinical Risk Score consolidates these factors into prognostic elements that can be used to select patients for hepatic metastasectomy. Patients who have 0 to 2 of these high-risk factors have 5-year survival rates of 40% to 60%, whereas patients with a score of 3 or higher have less than a 25% chance of surviving beyond the same time interval. Because the data to construct these scoring systems were generated before the advent of modern combination chemotherapy, the clinical usefulness of this model has been debated. Retrospective data suggest that tumor response to induction or “neoadjuvant” therapy as a marker of favorable biology may have more value than the aforementioned variables. Recently, multigene expression profiling of tumor tissue has been demonstrated to be highly predictive of overall survival in patients with resected hepatic colorectal metastases. Improved perioperative therapy, concentrating on minimizing blood loss, maintaining an appropriate and functional remnant liver volume, and maximizing hepatic regeneration and recovery through state-of-the-art surgical intensive care techniques, likely has contributed to the improved results in recent retrospective case series. Patients with normal hepatic and renal function can tolerate up to an 80% hepatectomy without significant perioperative mortality. The selective use of portal vein embolization to induce in vivo hepatic regeneration of the anticipated remnant liver is a technique that is predictive of the response to hepatectomy and can help certain patients who could tolerate complete metastasectomy. Portal vein embolization can be combined with neoadjuvant chemotherapy without compromising liver growth. Furthermore, our improved understanding of hepatic anatomy, physiology, and the factors responsible for regeneration has expanded the indications for operation and allowed more patients with stage IV disease to be considered for combination therapy. The ability to achieve a complete resection (i.e., R0 resection) with both gross and microscopically negative margins has been associated with improved recurrence-free survival. It is no longer necessary to perform anatomic or segmental resections on all patients with hepatic metastases to achieve disease control.
Role of Systemic Chemotherapy in Resectable Metastatic Disease
The most effective single agent as first-line therapy for metastatic CRC has historically been the fluoropyrimidine analogue, 5-fluorouracil (5-FU). For more than two decades, the synergistic combination of 5-FU and leucovorin or folinic acid (which inhibits thymidylate synthase) was the standard of care both as an adjuvant treatment for node-positive, resected colon cancers and as treatment for metastatic disease. The addition of oxaliplatin, a platinum-based alkylating agent, to the 5-FU–leucovorin backbone (FOLFOX) has increased the response rates and progression-free survival with lower rates of nephrotoxicity and myelosuppression. In some studies, FOLFOX has resulted in an overall survival benefit in the metastatic setting. Irinotecan is a topoisomerase inhibitor that when combined with 5-FU and leucovorin (FOLFIRI) is superior to 5-FU alone and comparable to FOLFOX. Capecitabine, an oral fluoropyrimidine antimetabolite with similar efficacy to 5-FU, is an equivalent, orally administered substitute for 5-FU in most regimens. Several biologic agents have been added to these combination regimens in an attempt to improve survival in patients with metastatic disease. Bevacizumab, a recombinant monoclonal antibody that blocks the activity of vascular endothelial growth factor A (VEGF-A), has proved of benefit in extending survival in patients with metastatic disease and is frequently added to the FOLFOX and FOLFIRI regimens. Cetuximab and panitumumab are monoclonal antibodies that block the epidermal growth factor receptor (EGFR) pathway. These anti-EGFR agents have efficacy against primary tumors that lack activating mutations in the KRAS gene (wild-type gene variant).
The surgeon treating patients with metastatic CRC must be familiar with safety profiles and potential deleterious effects of these regimens. The extended use of these agents (usually >6 to 12 weeks) may be associated with higher rates of postoperative complications and liver insufficiency. Irinotecan has been associated with both drug-induced steatosis and steatohepatitis, which may lead to increased morbidity and mortality. Oxaliplatin may produce sinusoidal dilatation, which can also increase the risk of postoperative complications. The rare, but potentially lethal side effects of bevacizumab include hypertension, increased risk of arterial thromboembolism, gastrointestinal bleeding, and perforation. Its antiangiogenic effects and long circulating half-life (approximately 6 to 8 weeks) have been associated with delayed wound healing. Most physicians interrupt bevacizumab therapy at least 4 to 6 weeks before surgical metastasectomy. Anti-EGFR agents, in particular cetuximab, have not been associated with the same risk profile or increase in postoperative morbidity or mortality.
Patients with unresectable CRC liver metastases who are treated with modern systemic chemotherapy alone have a median survival of 22 months, compared with the dismal natural history of the disease in untreated patients (median survival of 6 to 12 months). Several randomized controlled, multicenter trials have established the efficacy of the current standards of care for the treatment of metastatic disease.
The rate of complete pathologic response to chemotherapy has been estimated to be less than 5%. In addition, there is a poor correlation between complete radiologic and pathologic responses; residual disease (documented by pathologic examination of the resected liver) is present in more than 80% of lesions that disappear on high-resolution cross-sectional and metabolic imaging studies. Surgical metastasectomy alone can yield disease-free survival rates in the range of 20% and 5-year survival rates of 25% to 40%. Unfortunately, recurrence after resection is common. In approximately 50% of these cases, the recurrence occurs within the liver, and it is the only site of disease. Long-term disease-free survival is unlikely without a combination of complete metastasectomy and modern chemotherapy. The proper sequencing of surgery and chemotherapy remains unclear. The European Organization for Research and Treatment of Cancer (EORTC) trial 40983 (EPOC), a randomized, controlled phase III trial of perioperative FOLFOX versus surgery alone in patients with resectable hepatic CRC metastases, showed a significant improvement in progression-free survival from 28% to 36%, with an objective tumor response rate in all treated patients of 43%. Postoperative complications were more frequent in the chemotherapy group, although they were reversible and there was no increase in mortality. The median overall survival was 61 months in the chemotherapy group versus 54 months in the surgery-only group, which was not statistically significant, although the study was likely underpowered to detect a difference and only 63% of patients in the chemotherapy group received treatment post resection. A trial attempting to determine the optimal timing of systemic treatment (NSABP C-11, postoperative chemotherapy vs. preoperative and postoperative chemotherapy) was terminated due to poor accrual. The impact of biologic agents on survival after resection of hepatic metastases is still uncertain. The addition of cetuximab to standard chemotherapy in patients with resectable tumors was shown to be detrimental to progression-free survival and was an unexpected finding in a randomized trial. A summary of the outcomes of current combined-modality treatment is shown in Table 134.1 . Many of these studies included patients who initially had unresectable disease and became resectable after chemotherapy. Along the same lines as the EORTC EPOC trial, there have been two meta-analyses evaluating the role of chemotherapy in patients with resectable liver metastases. Both studies found a benefit in disease-free survival but not for overall survival.
TABLE 134.1
Survival Outcomes in Patients With Metastatic Colorectal Cancer Treated With Modern Combined Chemotherapy and Resection
| Study | No. of Patients | Initially Resectable | Regimen | Disease-Free Survival | Overall Survival |
|---|---|---|---|---|---|
| EORTC 40983 phase III RCT (EPOC) | 152 151 |
Yes | Surgery FOLFOX + Surgery + FOLFOX |
28.1% 36.2% (3 yr) P = .041 |
47.8% 51.2% (5 yr) P = NS |
| Ychou et al. phase III RCT | 153 153 |
Yes | 5-FU + leucovorin FOLFIRI |
46% 51% (2 yr) P = .44 |
71.6% 72.7% (3 yr) P = .69 |
| Adam et al. | 701 | No | FOLFOX | NA | 34% (5 yr) |
| Wein et al. phase II trial | 20 | Yes | FOLFOX | 52% (2 yr) |
80% (2-yr DSS) |
| Taieb et al phase II trial | 47 | Yes | FOLFOX followed by FOLFIRI | 47% (2 yr) |
89% (2 yr) |
| Barone et al. | 40 | No | FOLFIRI | NA | 63.5% (2 yr) |
| Masi et al. | 196 | No | FOLFOX/FOLFIRI? | 29% (5 yr) |
42% (5 yr) |
| First-BEAT trial | 107 | No | Bev + 5-FU based | NA | 89% (2 yr) |
| N016966 study | 34 44 |
No | Placebo + XELOX/FOLFOX Bev + XELOX/FOLFOX |
NA | 82.3% 90.9% (2 yr) |
| New EPOC | 117 119 |
Yes | FOLFOX or XELOX Above regimen + cetuximab |
20.5 months 14.1 months (PFS) P = .03 |
NA |
BEAT , Bevacizumab Expanded Access Trial; Bev , bevacizumab; DSS , disease-specific survival; EORTC , European Organization for Research and Treatment of Cancer; FOLFIRI , 5-fluorouracil, leucovorin, and irinotecan; FOLFOX , 5-fluorouracil, leucovorin, and oxaliplatin; 5-FU , 5-fluorouracil; NA , not available or not reported; NS, not significant; PFS , progression-free survival; RCT , randomized controlled trial; XELOX , capecitabine and oxaliplatin.
There are potential advantages and disadvantages to the neoadjuvant approach. The theoretical advantages include eliminating micrometastatic disease, in vivo cytoreduction to reduce the amount of hepatic parenchyma required for complete resection, the ability to individualize the chemotherapeutic regimen to improve efficacy, and, most importantly, to select patients who may benefit from metastasectomy. The potential disadvantages are biologic or chemotherapy-induced toxicity, inducing a complete radiologic response making the lesion(s) difficult to identify intraoperatively, and missing a window of opportunity to cure patients with resectable lesions that may progress on therapy and become unresectable.
Synchronous Versus Metachronous Disease
Fifteen percent to 25% of CRC patients have synchronous liver metastases at the time of diagnosis. In most cases, metastatic disease is recognized within the first year (or >6 months) after diagnosis. The biologic importance of synchronous metastatic disease is controversial. One hypothesis is that there may be no significant difference between synchronous and metachronous metastatic disease (i.e., lead-time bias). Several studies suggest that the presence of synchronous disease may be associated with a more adverse prognosis secondary to more aggressive tumor behavior (i.e., higher incidence of bilobar liver and extrahepatic metastases). The optimal treatment of synchronous metastases has been debated. Most patients are treated with a limited period (2 to 3 months) of systemic chemotherapy and are restaged. Surgical resection may be appropriate in the absence of disease progression and if both the primary and all sites of metastatic disease can be resected with acceptable morbidity and mortality.
The concept of staged resection has evolved in response to retrospective data that suggest higher rates of morbidity and mortality with combined resections. There are now multiinstitutional, retrospective data to suggest that simultaneous resections are not only feasible, but may be advantageous in selected cases when performed in high-volume centers. More recent reports have shown that morbidity and mortality are not increased in simultaneous hepatic and colorectal resections, even when major liver resections are performed. Simultaneous resection can translate to decreased total hospital stay, lower costs, and, in some single-institution reports, lower total complication rates. In an international multicenter study, postoperative morbidity was 20% and mortality was 3%, with no increased risk of complications when comparing staged or simultaneous approaches. In a separate report with long-term follow-up, the oncologic outcomes were also found to be similar with both approaches. These are retrospective reports, and simultaneous resections were more frequently performed in patients with fewer or smaller metastases, with more proximal tumors, and better prognostic features. When synchronous hepatic metastatic disease is present in the setting of the primary tumor located in the rectum as opposed to the colon, the timing of systemic therapy and local regional therapy (i.e., radiation therapy) may well be the most critical factor in determining the sequencing of hepatic resection and resection of the rectal primary. In the absence of evidence-based data, most treatment decisions are individualized based on tumor biology, patient factors (e.g., performance status, liver health), and the location and distribution of metastatic disease.
Liver Resection in the Presence of Extrahepatic Disease
Patients with CRC liver metastases and EHD have traditionally not been considered as candidates for metastasectomy. The improved efficacy of chemotherapy has allowed some to reconsider the role of surgical resection in these individuals. The most prudent approach involves a limited period of neoadjuvant chemotherapy to allow patients with favorable tumor biology to be selected for aggressive surgical intervention.
Five-year survival rates of up to 20% to 30% have been reported in patients with EHD. Two of the largest series of patients undergoing liver resection and extrahepatic metastasectomy to date reported a 5-year predicted survival of 27%, which is consistent with the previous reports with smaller sample sizes. The majority (84% to 95%) of patients in these series eventually developed recurrence, emphasizing that liver resection in the presence of EHD is very often not curative. A systematic review of the outcomes of resection for patients with synchronous liver and EHD reported similar results in terms of survival.
The value of removing metastatic disease in “distant” regional lymphatic basins (e.g., periaortic or hepatic pedicle) remains controversial. It is likely that this clinical situation reflects adverse tumor biology associated with poor long-term survival. Reports of hepatic resection and aggressive portal lymphadenectomy in combination with chemotherapy have been associated with 5-year survival rate approaching 20% when lymph node metastases are limited to the portal basin, with no long-term survivors found among patients with celiac or retroperitoneal metastases.
The evolution of hyperthermic intraperitoneal chemotherapy in combination with cytoreductive surgery offers another possible aggressive treatment for selected patients with peritoneal disease. The value of this treatment in the setting of concomitant solid-organ metastases is unknown.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here