In their article in this issue of ONCOLOGY, authors Bartlett and Chu discuss a very provocative suggestion that the possibility exists to cure patients with advanced colorectal cancer. Before we can discuss whether or not we can cure stage IV colorectal cancer, however, we first need to decide what defines cure. By convention, the authors have focused primarily on 5-year survival as their survival endpoint, largely because the majority of studies report their outcomes utilizing this timeline. However, a certain percentage of patients will recur even after the 5-year time point. In order to really define “cure,” we need to follow patients long enough to realize the asymptotic portion of their survival curve. The group from Memorial Sloan-Kettering Cancer Center (MSKCC) reported on their long-term 10-year follow-up of 612 patients who underwent hepatic resection for colorectal cancer liver metastases and found that the true flat portion of the curve occurred at the 10-year time point.[1] There was still attrition of patients due to recurrence between 5 and 10 years after hepatic resection, as one-third of patients suffered a cancer-related death during this time period. After 10 years, however, very few patients recurred; only 1 patient out of the 102 still alive at 10 years experienced a disease-specific death, and the survival curve was nearly a flat horizontal line. Thus, based on these data, we can “cure” patients with metastatic colorectal cancer, but this ultimate goal unfortunately is realized less than 20% of the time when we embark on treatment with curative intent.
In order to realize a balance that optimizes the risk-to-benefit ratio in favor of the patient, we have developed various risk-stratification scoring systems to help “select” the best patients to undergo an aggressive treatment regimen that includes resection.[2] Multiple factors, such as the number of liver metastases, size of lesions, characteristics of the primary tumor, and disease-free interval have all been used to select patients. No single scoring system, however, has been uniformly validated across independent patient populations from different institutions. In fact, in the series referenced from MSKCC, a substantial number of patients who were cured of their disease at 10 years originally had poor prognostic factors such as synchronous disease (7%), a disease-free interval less than 1 year (36%), bilobar disease (25%), a node-positive primary tumor (50%), multiple liver tumors (39%), and large hepatic lesions (35%).[1] Thus, despite our best efforts to select the ideal patients for aggressive surgery, we just simply do not have all the answers and ultimately, cancer biology still evades our complete understanding. Sometimes we are pleasantly surprised with patient outcomes and sometimes horribly disappointed. We must be aggressive in order to help that select minority of patients realize their potential for cure.
As we continue to push the envelope for resection of colorectal cancer liver metastases, the perioperative strategy more frequently is inclusive of techniques such as a two-stage resection approach, portal vein embolization, the reverse treatment strategy of removing the liver metastases prior to the primary tumor, and resection combined with intraoperative ablation. A projected narrow margin should not necessarily preclude an attempt at resection, as investigators have demonstrated similar outcomes between a 1 mm vs a 1 cm resection margin.[3] The primary consideration that must be maintained at the forefront of these aggressive treatment strategies is patient safety. One must never overlook the importance of an adequate future liver remnant, as what is left behind is always more important than what is taken out.
Systemic therapy with cytotoxic chemotherapy and/or biologics is usually integral to the therapeutic plan for patients with stage IV colorectal cancer. The European Organisation for Research and Treatment of Cancer (EORTC) trial[4] was the first completed randomized trial that independently demonstrated a survival advantage with a combination approach of surgery with chemotherapy vs surgery alone. The actual contribution, however, of chemotherapy to improved survival was modest at best. The difference between the two groups in 3-year progression-free survival was only 7.3%, favoring the perioperative chemotherapy group. This result often has been misinterpreted and wrongly applied to the clinical arena, with the conclusion that all patients should receive chemotherapy before undergoing resection. This is not valid, since trial design only compared combination therapy vs surgery alone, not the order in which chemotherapy should be administered. A common rationale for administering chemotherapy prior to resection for patients with resectable disease is to assess disease biology and identify patients who may not benefit from resection. In the EORTC trial, only 7% of patients in the perioperative chemotherapy group progressed during treatment and were excluded from subsequent resection. Thus, with the current regimens of FOLFOX and FOLFIRI having such good response rates in the initial first-line setting, the actual demonstration of bad tumor biology during the early treatment period occurs in a small minority of patients. Furthermore, the potential for increased postoperative morbidity associated with preoperative chemotherapy must be considered when devising a treatment strategy. This author is not saying that chemotherapy should not be utilized, but rather that chemotherapy is more of an adjunct treatment to surgery, rather than vice versa, for patients who have resectable disease.
The next advance for treating patients with advanced colorectal cancer should be further personalization of therapy. This has become the goal of treatment across all different types of cancers, and advanced colorectal cancer is no different. The relationship of cetuximab(Drug information on cetuximab) efficacy with KRAS mutation status is one such example,[5] but we need to go deeper. Efforts must be made to discover molecular biomarkers and genetic profiles of tumors that will not only prognosticate outcomes, but also predict response to a certain treatment regimen or strategy.[6] The prognostic and predictive values of a biomarker are often intertwined, and randomized prospective trials are needed to tease out the true predictive utility. Is personalized therapy ever going to be a reality? This remains to be seen, but encouraging small steps in the right direction have been observed in different disease types. Our understanding of cancer biology must continue to evolve and mature in order for us to realize the hope of cure for our patients.
Financial Disclosure: The author has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
