Systemic chemotherapy for colorectal liver metastasis: Impact on surgical management

Introduction

Over the past decades, the management of colorectal liver metastases (CLM) has been transformed by several improvements in both medical and surgical fields (see Chapters 90 , 97 , 101 , and 102 ). The development of modern and efficient chemotherapeutic regimens has increased tumor response rates as well as overall prognosis of metastatic patients. A better knowledge of tumor biology has allowed tailoring indications of targeted therapies, thus initiating the era of personalized medicine. From the surgical point of view, technical refinements have pushed the boundaries of resectability while maintaining acceptable surgical risk.

The interplay between surgery and chemotherapy has become the cornerstone of the modern management of CLM, and this “onco-surgical approach” is now recognized as a standard. Therefore understanding the consequences related to chemotherapy is essential to define the most appropriate onco-surgical strategy.

The aim of this chapter is to discuss the impact of systemic chemotherapy on the surgical management of patients with CLM. We discuss the role of chemotherapy in three different clinical situations:

Initially resectable disease
Initially unresectable disease with a possibility of resectability in case of good response to conversion chemotherapy
Definitively unresectable disease

A brief history of systemic chemotherapy in colorectal liver metastases

The spontaneous median survival of a metastatic colorectal cancer (CRC) without treatment is 5 months. The past two decades have seen the development of new systemic chemotherapies, reaching around 3 years of median survival with the most recent protocols.

In 1957, 5-fluorouracil (5-FU) was the first chemotherapy authorized in colorectal cancer, but when used alone, its effect was limited, offering a response rate of around 10%. Response rates were doubled after discovering that folinic acid (leucovorin) stabilized the interaction between 5-FU and thymidylate synthase, thereby increasing the effect of 5-FU. ^, The modalities of administration (bolus, continuous infusion) were later improved to decrease toxicity without impairing tumor control. De Gramont et al. proposed the LV5-FU2 protocol (biweekly cycles consisting of an intravenous [IV] administration of leucovorin over 2 hours followed by a bolus of 5-FU and a slow IV administration of 5-FU over 2 days), which quickly gained wide acceptance.

The 2000s were marked by the introduction of two major anticancer drugs effective in colorectal cancer: irinotecan and oxaliplatin. Irinotecan (a topoisomerase I inhibitor, transformed after a first hepatic passage into an active metabolite SN38 by a carboxylesterase) demonstrated a benefit in terms of survival in patients resistant to 5-FU. Associated with LV5-FU2, Douillard et al. showed a higher response rate (49% vs. 31%) and a median survival reaching 17 months. This protocol was later simplified to become FOLFIRI (see Chapter 97 ).

Oxaliplatin is a derivative of platinum salts, which inhibit DNA repair. The first report concerning the high response rate observed with this drug was made in 1990. Then the combination of LV5-FU2 and oxaliplatin (FOLFOX) became a standard by showing better oncologic results compared with LV5-FU2 alone. The sequence FOLFOX followed by FOLFIRI, in the event of resistance, was compared with the reverse sequence in a European trial. The results did not show differences in survival, suggesting that the sequence of the two protocols and their effectiveness were comparable.

The therapeutic armamentarium in metastatic CRC was enriched in 2004 with the development of two monoclonal antibodies directed against an oncogenic activation receptor present on the surface of tumor cells: cetuximab targeting the epidermal growth factor receptor (EGFR) and bevacizumab directed against the vascular endothelium growth factor (VEGF). ^, Both targeted therapies combined with FOLFOX or FOLFIRI regimens further improved outcomes in metastatic CRC patients. ^, It was later discovered that the efficiency of anti-EGFR was compromised in tumors exhibiting a mutation of KRAS, a protein involved in the transduction of the EGFR. This finding marked the starting point of tailored oncology in CRC. ^, Additional mutations of the RAS/RAF pathway conferring resistance to anti-EGFR therapies were later identified, and overall about half of metastatic CRC patients cannot be efficiently treated by anti-EGFR therapies. The efficiency of more aggressive protocols combining three chemotherapeutic agents (oxaliplatin, irinotecan, and LV5-FU2) with bevacizumab was demonstrated in 2010. This aggressive protocol yielded objective response rates of 65% and a median survival of 31 months. The end of the 2010s has seen the impressive results of immunotherapy in tumors with microsatellite instability (MSI-H, about 5% of patients). ^, A combination of immune checkpoint inhibitors makes it possible to obtain an objective response in 69% of MSI-H patients with resistant disease to systemic chemotherapy. The evolution of systemic treatment in metastatic CRC over the past decades across main trials is described in Table 98.1 .

TABLE 98.1

Main Randomized Trials of First-Line Chemotherapy in Metastatic CRC Since 2000

FIRST AUTHOR (STUDY NAME)	YEAR	NO. OF PATIENTS	RAS/RAF	EXPERIMENTAL ARM (VS. CONTROL ARM)	MEDIAN OVERALL SURVIVAL, * MONTHS	MEDIAN PFS, * MONTHS	OBJECTIVE * RR, %
De Gramont	2000	420		FOLFOX (vs. LV5-FU2)	16.2	9.0	50.7
Douillard	2000	387		FOLFIRI (vs. 5FU-folinate)	17.1	6.7	35
Hurwitz (AVG2107g)	2004	813		Irinotecan 5FU leucovorin (IFL)–bevacizumab (vs. IFL)	20.3	10.6	44.8
Falcone	2007	244		FOLFOXIRI (vs. FOLFIRI)	22.6	9.8	66
Van Custem (CRYSTAL)	2009	1198	KRAS	FOLFIRI-cetuximab (vs. FOLFIRI)	24.9	9.9	47.1
Bokemeyer (OPUS)	2011	337	KRAS-BRAF	FOLFOX4-cetuximab	22.8	8.3	67
Douillard (PRIME)	2013	512	KRAS-NRAS -BRAF	FOLFOX 4-panitumumab (vs. FOLFOX 4)	25.8	10.1	_
Schwartzberg (PEAK)	2014	285	KRAS	FOLFOX6-panitumumab vs. FOLFOX6-bevacizumab	41.3	13.1	64
Heinemann (FIRE-3)	2014	592	KRAS	FOLFIRI-cetuximab vs. FOLFIRI-bevacizumab	28.7	10.0	62
Loupakis (TRIBE)	2014	508		FOLFOXIRI-bevacizumab (vs. FOLFIRI-bevacizumab)	29.8	12	65
Venook (CALGB-SWOG80405)	2017	1074	KRAS	FOLFIRI or FOLFOX6-cetuximab vs. bevacizumab	30.0	10.5

* Results given for the experimental arms.

Evolution of the definition of resectability

The first patients who underwent surgical resection for CLM were very selected. Surgery was not considered beyond three lesions or in the presence of concomitant extrahepatic disease. With the combined improvements of surgery and chemotherapy, criteria of resectability have been expanded and the current definition is no longer dogmatic but rather pragmatic. The number of lesions and the maximal size of lesions are no longer absolute contraindications for resection. To date, resectability can be defined as the possibility to resect all lesions present at the time of diagnosis while preserving sufficient volume of functional remnant liver. Most groups agree that a safe volume of remnant liver should be at least 25% to 30% of the total liver functional volume (i.e., after subtracting tumor volume) and/or 0.5% of the remnant liver volume to body weight ratio.

Preoperative chemotherapy in patients with upfront resectable disease

Advantages and disadvantages of preoperative chemotherapy for upfront disease

Whether chemotherapy should be given before surgery in patients with upfront resectable disease remains a matter a debate. The advantages of such an approach are as follows:

1.
Preoperative chemotherapy makes it possible to test the chemosensitivity of tumors. This enables us to choose appropriate adjuvant chemotherapy and to select good surgical candidates, based on the evidence that progression while on chemotherapy impairs oncologic outcomes.
2.
Preoperative chemotherapy may also facilitate resection with tumor-free margins by tumor downsizing and the preservation of normal parenchyma.

Disadvantages include the following:

1.
It can be responsible for parenchymal injury of the nontumoral liver, thus increasing the morbidity of resection (see Chapter 69 ).
2.
The radiologic disappearance of small lesions (so-called “missing” metastases) is another drawback because missing lesions on imaging does not mean sterilization of the tumor, and their surgical removal is more challenging and may not even be possible.

A special situation is the one concerning patients with resectable metastases progressing while on chemotherapy. In patients with at least three lesions, 5-year overall survival after hepatectomy was only 8% in patients with progressive disease versus 37% and 30%, respectively, for patients with responding or stable disease, indicating that progression while on chemotherapy should be considered as a relative contraindication for resection. A later analysis from the LiverMetSurvey registry enabled researchers to refine the indications for hepatectomy in a context of tumor progression on chemotherapy. It appears that patients with low CEA (<200 ng/mL) and no more than three lesions, with none of them being larger than 5 cm, experienced 5-year overall survival rates of 53%. This favorable outcome argues in favor of resection in selected patients despite progression.

Upfront resection versus neoadjuvant chemotherapy: Evidence from the literature

A randomized trial has compared perioperative FOLFOX versus no chemotherapy in patients with upfront resectable CLM with no more than four lesions. Patients treated by chemotherapy experienced a significantly improved disease-free survival at 3 years (25% vs. 16%) at the cost of a higher morbidity. However, this advantage in disease-free survival did not improve overall survival. The main limitation of this trial was the absence of adjuvant chemotherapy in the control group. Later, three randomized controlled trials (PANTER NCT01266187, EXPERT UMIN000007787, NCT01035385) have attempted to compare perioperative chemotherapy versus upfront surgery followed by adjuvant chemotherapy, but all have been closed due to low recruitment issues. A retrospective analysis based on the LiverMetSurvey registry showed that preoperative chemotherapy did not yield better results in patients with a single, metachronous metastasis not exceeding 5 cm.

More recently, the strategy “upfront surgery plus adjuvant chemotherapy” was compared with perioperative chemotherapy in patients with upfront resectable liver disease. After adjustment for confounding factors, there was no difference in disease-free survival or overall survival. So far, preoperative chemotherapy does not seem to provide an obvious survival benefit, provided that efficient chemotherapy is given postoperatively. Therefore upfront surgery can be reasonably considered in patients with a limited metachronous disease requiring a low-risk hepatectomy. In contrast, synchronous disease is recognized as a more advanced disease, and in a consensus meeting, most experts agreed to propose preoperative chemotherapy even when disease was resectable. By combining number and maximum size of metastases in the large cohort of the LiverMetSurvey registry, we concluded that the more extensive the disease, is even when initially resectable, the more useful the preoperative chemotherapy is in terms of survival expectancy after liver resection (unpublished data). In our own practice, we consider the use of chemotherapy in case of synchronous metastases and/or in patients with more than three metastases, mainly when the maximum diameter exceeds 3 cm.

Type of chemotherapy in upfront resectable patients

The two trials (EORTC intergroup trial 40983, new EPOC) that addressed the question of perioperative chemotherapy in patients with resectable CLM have chosen the FOLFOX regimen. However, FOLFIRI was allowed in patients previously treated by adjuvant FOLFOX after resection of the primary tumor in the new EPOC study. Hence, no other regimen can be recommended in patients with resectable disease. The new EPOC study evaluated the effect of adding cetuximab to FOLFOX in wild-type RAS patients with upfront or suboptimally resectable disease. Surprisingly, the addition of cetuximab to perioperative FOLFOX in a population of wild-type RAS patients was associated with poorer progression-free survival and lower overall survival. ^, Differences in tumor biology, selection of a more aggressive disease by cetuximab, and increased number of missing metastases have been advocated to explain this unexpected finding. In any case, the use of cetuximab in resectable patients should be avoided.

Patients with initially unresectable but potentially resectable disease

Increasing evidence has shown that the prognosis of patients in whom liver disease can be resected is far better than for those treated by chemotherapy only ( Fig. 98.1 ). In addition, patients with initially unresectable disease who were switched to be resected after a major response to chemotherapy have experienced almost similar outcomes as those with upfront resectable disease ( Fig. 98.2 ). These findings have introduced the concept of “conversion chemotherapy” and have led to considering secondary liver resection as a primary endpoint in the strategy of treatment of patients with nonresectable metastases. Therefore the aim of preoperative chemotherapy should be interpreted according to the initial resectability of the liver disease, which highlights the need for a multidisciplinary team approach and a careful assessment of resectability by experienced liver surgeons.

FIGURE 98.1, Kaplan-Meier survival according to resection in patients who underwent surgery for colorectal liver metastases in the LiverMetSurvey registry.

FIGURE 98.2, Illustration of onco-surgical management of a patient with initially unresectable colorectal liver metastases (CLM).

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here