Defining Resectability: From the Surgeon Perspective to the Multidisciplinary Approach
Surgical resectability of CLM have been historically interpreted as an exclusively technical problem: the initial criteria proposed for the definition of resectability were derived from the analysis of surgical series aiming to identify tumor characteristics associated with the best outcome in terms of survival, rather than to comprehensively assess the complexity of the patient-disease interaction. Moving from the classical surgical criteria, a paradigm shift has recently been completed and surgical resectability of CLM is now recognized as a many-sided scenario requiring a multidisciplinary approach by a highly trained team.
In 1989, Steele and colleagues identified several tumor lesions characteristics that should have guided the surgical approach to CLM in order to obtain the best outcome: the number of lesions (less than four, in the same lobe), the maximum lesion dimensions (less than 5 cm), the timing of metastases (metachronous), the absence of metastatic spreading outside the liver, an adequate (more than 1 cm) margin of healthy liver tissue for all resected lesions, an adequate liver remnant after resection and a radical (R0) resection of all CLM. Since then, several authors have questioned the relevance of many of these indications. With regards to the number and dimensions of CLM, it is now clear that strict limitations cannot be identified, since important surgical series have clearly demonstrated that prognosis does not differ significantly between patients with one to three CLM and patients with four or more CLM, and that lesion dimensions per se do not represent a contraindication to resection, but should both be integrated into more comprehensive prognostic scores. In addition, the synchronous appearance of CLM is no longer considered an absolute contraindication to surgery or a major determinant of poor prognosis in this setting, particularly in a more recent series of patients treated with more active combination chemotherapy.
The presence of extrahepatic disease has intuitively raised perplexity about the benefit of surgical resection of CLM in the past: it seemed obvious that any local procedure on liver lesions could have not been successful without concomitant treatment of distant disease in other organs. However, improvements in systemic medical treatment, while expanding the possibility of liver locoregional procedures, have also made surgical removal of extrahepatic disease possible and, theoretically, increased the chances of achieving complete pathologic response not only at the sites of micrometastatic spread, but also in limited established extrahepatic metastases, which will be not resected at all. This concept has been proven valuable, at least for metastases in the lung or in the hepatic pedicle lymph nodes, with 5-year OS rates in the range of 25 to 30% in selected series from high volume centers.
Therefore, what are the current limitations for liver surgery of CLM? According to the most recent statements of surgical consensus conferences, the only relevant rules to be followed when performing surgical resection of CLM are represented by the achievement of a complete (R0) resection, the preservation of two contiguous liver segments (with adequate vascular inflow and outflow) and adequate future liver remnant (more than 20% of the total volume in a healthy liver). Therefore, in the last 20 years, the only main factors surviving the seismic shift in the onco-surgical approach to CLM is represented by the possibility of achieving a R0 resection, independent of the extent of the tumor to be removed, while carefully considering the healthy liver tissue that will remain after the operation. It is worth noting that R0 resection is defined independently of the surgical margin width, which can be less than 1 cm with no impairment in survival, provided that all of the tumor is removed. Moreover, in the event that the remnant liver will be not sufficient after the planned surgical procedure, innovative approaches, such as preoperative portal vein embolization (resulting in subsequent hypertrophy of future liver remnant) and staged hepatic resection may offer the possibility to reconsider a larger percentage of mCRC patients for metastasectomy.
The current CLM management is thus centered around the concept of achieving complete tumor resection, but even though all surgeons agree on the aim of the surgery strategy, the consensus regarding the feasibility of liver resection in single cases still suffers from considerable disagreement. One of the most interesting experiences in the field is that reported in the Cetuximab in Neoadjuvant Treatment of Non-Resectable Colorectal Liver Metastases (CELIM) trial, which will be discussed in more detail later in the article. The review of resectability for the 106 patients, in whom paired baseline and follow-up scans were available during the study, was based on radiological images alone and involved seven surgeons. Investigators confirmed that, after chemotherapy, a higher number of patients were considered resectable (60 vs 32%; p < 0.0001); they also confirmed that objective response to chemotherapy had a significant effect on change of resectability. The voting pattern of surgeons demonstrated considerable inter-individual variation, with identical votes between surgeons noted in 64.5% of the cases; however, the occurrence of completely opposed views (i.e., one surgeon classifying a lesion as resectable and another as nonresectable) was rare (6.8%).
These results underline that, even if resectability is still surgically defined (R0 resection), a more accurate evaluation of the potential value of resection in the history of the disease is needed in single cases; resectability as a synonym of technical feasibility can no longer be accepted as the guide for treatment decision-making. Moving further towards a deeper definition of resectability, we should be able to integrate prognostic evaluation in such a merely technical definition, as the outcome of radically resected patients widely varies according to a multitude of clinical and molecular factors. As shown by Tomlinson et al., 17% of patients undergoing radical resection of metastases are alive at 10 years, demonstrating that the majority of resected patients is not definitively cured. As will be discussed in detail in the following section, validated clinical scores are already available, while molecular markers of outcome are still under investigation. As insights into the biology of CLM are provided in the near future, it will become clearer that resectability, far from being an exclusively surgical problem, is defined by the framework of a multidisciplinary strategy centered around a comprehensive patient and tumor evaluation, thus shifting from what is merely technically feasible to what is rational from an oncological perspective.
At present, using the aforementioned data, we can identify three different patient categories in mCRC: comprised between the two extremes represented by patients with easily resectable disease (i.e., those cases with a few CLM, which can be easily resected by the surgeons in the absence of poor prognostic indicators) and patients with unresectable disease who are unlikely to ever become resectable (owing to extensive or ill-located tumor burden that would be unresectable even in the case of a response to treatment), resides the vast majority of mCRC patients with initially unresectable disease owing to technical difficulties and/or the presence of poor prognostic factors. In balancing the relevance of the surgical and medical approaches in each patient subgroup, a clear definition of the aim of treatment is essential from the onset in order to establish a personalized therapeutic strategy. Resection is the only means of achieving a cure for mCRC but, for the majority of resected patients, it will result at best in long-term survival. Therefore, an optimal integration between medical treatment and surgery is essential in order to maximize the benefit of these different approaches according to the reasonable goal of loco-regional procedures in a systemic disease such as mCRC. As reported in the next sections, these new concepts have paved the way for prospective trials testing more modern regimens and biological agents in selected mCRC patients judged potentially resectable from a surgical point view but harboring different poor prognostic factors. Results of these studies are extremely useful for deriving information about different treatment options in different subsets of patients according to an accurate prognostic evaluation.