Prognostic Factors of Radiofrequency Ablation plus Systemic Chemotherapy for Unresectable Colorectal Cancer with Liver Metastasis

Introduction Survival outcomes in patients with unresectable colorectal cancer (CRC) liver metastasis treated by radiofrequency ablation (RFA) combined with systemic chemotherapy and correlation with potential prognostic factors were investigated. Material and Methods. A retrospective cohort study was conducted on 61 CRC patients with unresectable liver metastasis who underwent liver tumor-directed percutaneous RFA combined with conventional systemic chemotherapy between October 2013 and September 2018. Survival analyses were conducted using the Kaplan-Meier method, and the log-rank test was used to characterize differences in the median survival time and the 1-year, 3-year, and 5-year overall survival rates of subgroups to identify prognostic factors. Results Median overall survival and progression-free survival of all patients were 32 and 14 months, respectively. The cumulative survival rates at 1-, 3-, and 5-years were 93.2%; 44.5%, and 38.2%, respectively. Univariate analysis revealed that pre-RFA serum CEA levels, Eastern Cooperative Oncology Group (ECOG) status, number of liver lesions, the size of the largest lesion, and the total lesion size were prognostic factors. However, multivariate analysis demonstrated that only the number of liver lesions and the size of the largest lesion were independent prognostic factors for survival. Conclusion RFA plus systemic chemotherapy provides an encouraging survival outcome for patients with unresectable CRC liver metastasis. Multivariate analysis demonstrated that the number and size of liver metastatic lesions are independent prognostic factors for survival.


Introduction
Colorectal cancer (CRC) is the third most common cancer by incidence and the second leading cause of cancer-related deaths in both sexes [1]. The incidence of the disease in Vietnam has increased recently [2]. Synchronous liver metastasis reportedly occurs in approximately 20-25% of CRC patients at the time of initial diagnosis. Moreover, 20-30% of patients progress to experience liver metastasis within the first three years after CRC diagnosis. Liver metastasis is considered the main cause of death in CRC patients [3,4].
Systemic chemotherapy is the standard of care for CRC patients with liver metastasis. However, liver metastatic lesions typically respond poorly to chemotherapy alone; thus, survival outcomes are suboptimal despite the recent introduction of novel chemotherapeutic agents and targeted therapies. Although liver resection is the most curative treatment for liver metastasis, only about 20-25% of CRC patients with liver metastasis are eligible for resection. Radiofrequency ablation (RFA) is an alternative nonsurgical treatment for CRC liver metastasis when the liver metastatic lesions are not eligible for resection or because of comorbidities.
However, data on the efficacy of RFA for the treatment of CRC liver metastasis is still limited. Several clinical studies demonstrated that RFA is an effective and safe treatment for liver metastasis in CRC; however, local and distant recurrent rates remain high. The combination of RFA and systemic chemotherapy has also been evaluated in some clinical studies and resulted in promising outcomes for CRC liver metastasis as demonstrated by progression-free survival (PFS) and overall survival (OS). However, the treatment outcome of treated patients is still heterogenous; we are still in the lack of given factors that can be used for prognosing the responses of patients [5][6][7][8][9][10][11][12][13][14][15][16]. Therefore, the aim of our study was to evaluate survival outcomes and some potential prognostic factors in patients with CRC liver metastasis treated by using RFA combined with systemic chemotherapy.  Helsinki (1975) and was approved by the Institutional Review Board of 108 Military Central Hospital (O-62720149). As this study was retrospective in nature, written consent from the patients was not required.
Patients were excluded if they had previous liver interventions or surgical liver resection, portal vein thrombosis, presence of active gastrointestinal (GI) bleeding, and severe coagulation abnormalities (prothrombin < 60%, platelet count < 50 G/l). Pregnant or feeding women were also excluded.
All patients were discussed in a multidisciplinary tumor board that included medical oncologists, liver surgeons, radiologists, and hepatologists who performed RFA. The number and the location of the liver lesions, the presence or absence of extrahepatic metastasis, and comorbidities were taken into consideration when determining treatment approaches. Patients deemed to have resectable liver metastasis with severe comorbidity were considered unsuitable for surgery.

RFA Procedure and Chemotherapy
Regimens. The Cooltip RF Ablation System E Series (USA) with multiple probe approaches was used for the RFA technique and performed by two hepatologists who had more than 5 years of experience in percutaneous local ablation for primary and secondary liver cancer. The intervention was conducted percutaneously with ultrasound guidance, and all patients were given local and intravenous anesthesia. Each lesion was calculated to be ablated with at least a 5 mm ablative margin. When lesions were located in difficult locations (i.e., near gastrointestinal tract), an artificial-ascites RFA procedure was used to avoid complications. Tumor necrosis was evaluated by contrast-enhanced CT scan 2 weeks after intervention. A second additional RFA was conducted in patients with incomplete tumor ablation. Systemic chemotherapy was given to patients 2-4 weeks after RFA intervention when all RFA-related adverse events had resolved, and complete ablation of liver lesions was confirmed. FOLFOX-4 or FOLFIRI regimens were chosen for each patient depending on which systemic treatment the patient received for the primary cancer (FOLFOX-4 for first line and FOLFIRI for second line chemotherapy).
Detail of chemo regimens [13]: All patients were treated with pre-and postchemotherapy antiemetics, including 5-HT3 blockers and dexamethasone.
Patient tolerance to chemotherapy was evaluated at the time of the next cycle to adjust the dose of chemo drugs. Follow-up images and serum CEA levels were obtained every 3 months after the first chemotherapy cycle to assess recurrence or progression. Patients with evidence of any recurrence or progression would receive further treatments or just palliative care, after being reevaluated by a multidisciplinary tumor board.

Primary Endpoint.
The primary endpoint of this study was OS, defined as the date of RFA for liver metastases to the date of death. Patients lost to follow-up were censored. The second endpoint was progression-free survival (PFS), defined as the date of RFA to the date of disease progression or death.

Statistical
Analysis. Baseline characteristics were described as numbers (%) and mean with standard deviation. Clinical variables included age, gender, ECOG, primary cancer, type of metastasis, number of liver metastatic lesions, size  (2) progression-free survival was 14 months (95% CI: [11][12][13][14][15][16][17]. The cumulative survival rates at 1-, 3-, and 5-year follow-up were 93.2%; 44.5%, and 38.2%, respectively. The median survival time and survival rates of subgroups are presented in Table 2. Univariate analysis revealed that the number of liver lesions, the size of the largest lesion, the sum of lesion diameter, and pre-RFA ECOG status were prognostic factors for overall survival (Figure 1). Patients with 1-3 liver lesions had an improved median OS, compared with patients who had 4-5 liver tumors (41 months vs. 30 months, p = 0:008). Patients in whom the size of the largest liver lesion was <3 cm also had better median OS, compared with those in whom the largest liver lesion was >3 cm (38 months vs. 30 months, p = 0:026). Significant differences in median OS was also demonstrated based on ECOG status and levels of serum CEA (p < 0:05). Notably, age, sex, primary cancer, type of metastasis, and chemo regimen criteria did not correlate significantly with patient survival after treatment. A subsequent multivariate analysis revealed that the number of liver lesions (hazard ratio [HR] for 1-3 vs. 4-5, 2.91; confidential interval, 1.1-7.4; p = 0:02) and the size of the largest liver lesion (hazard ratio [HR] for ≤3 cm vs. 3-5 cm, 4.72; confidential interval: 1.54-14.4; p = 0:01) were independent predictors of OS (Table 3).

Discussion
Although this was not a randomized study, the survival outcomes of the CRC patients with liver metastasis reported here  [15]. Survival of patients in the EORTC 40004 study, which was better than survival in the study being described here, may be explained by the addition of bevacizumab to systemic treatment and longer follow-up. Bevacizumab is an expensive therapy and not fully covered by government reimbursement; therefore, it is currently not routinely used in clinical practices [16,17]. Better survival outcomes were also seen in a study of Solbiati et al. with a median overall survival of 53 months, but the patient population included only small, favorably located liver metastasis (i.e., maximum diameter of any metastasis < 4 cm, each >1 cm away from hepatic hilum or GI tract) with no extrahepatic disease, and a proportion of these patients was eligible for resection [18]. Meanwhile, another prospective study by Berber et al. in the United States, involving 135 patients with CRC liver metastasis treated by laparoscopic RFA combined with systemic chemotherapy, demonstrated a median overall survival of only 28.9 months, shorter than that in our study. Berber's study recruited older patients (mean age 62 years old), more patients with previous chemotherapy exposure (108/135), and patients with extrahepatic disease, which may explain the outcome [19]. Patient survival in the current study was also higher than that in a study conducted by Gillams et al. (median survival of 27 months; 1-, 3-, and 4-year survival of 90%, 34%, and 22%, respectively). However, Gillams and colleagues included patients across a broad spectrum of disease severity: (1) did not exclude those with extrahepatic metastasis, (2) the number of liver lesions was 1-16, and (3) the diameters of lesions were 1-8 cm [20].  , and the presence of extrahepatic disease did not affect survival [19]. Shady et al. evaluated factors affecting outcomes of percutaneous RFA of CRC liver metastasis in a retrospective single-center study with 10 years of follow-up; using univariate analysis, the authors demonstrated that the size of the largest liver lesion (greater or less than 3 cm), serum CEA level (greater or lower 30 ng/ml), and presence or absence of extrahepatic diseases were independent predictors for overall survival. However, multivariate analysis revealed that only lesion size greater than 3 cm and more than one site of extrahepatic diseases was independent predictors of shorter overall survival [21]. In the analysis reported here, no extrahepatic spreads or lymph node metastases were present, an observation confirmed by chest/abdominal CT scan or whole-body PET/CT. Seifert and Morris investigated potential prognostic factors in 116 patients with CRC liver metastasis undergoing cryotherapy and reported that low serum CEA levels (<5 ng/ml), small (≤3 cm) diameter of liver metastasis, absence of untreated extrahepatic disease at laparotomy, absence of nodal involvement at primary resection, complete cryotherapy, synchronous development of liver metastasis, and good or moderate differentiation of primary tumors were independently associated with a favorable outcome [22]. The liver-tumor burden impacts survival in this study, an observation that is corroborated by other studies. This can be explained by the risk of microscopically incomplete ablation in patients with higher liver tumor burden. We should note that the liver tumor burden is also a very important risk factor for survival in patients with CRC liver metastasis after resection [23,24].
This study has some limitations. The small patient population and the retrospective nature of the study are potential drawbacks in that it might be insufficient to guide decisive conclusions on the other potential prognostic factors tested. However, the results of our study suggest that at least two parameters (i.e., total lesion number and size of the largest lesion) should be taken into account when considering RFA plus systemic chemotherapy for CRC patients with liver metastasis. The integration of these metastatic tumor characteristics in predicting the long-term survival of patients with CRC liver metastasis should also be addressed.
In conclusion, RFA plus systemic chemotherapy demonstrates an encouraging survival outcome for patients with unresectable CRC liver metastasis. The number of liver metastatic lesions ≤ 3 and size of the largest lesion ≤ 3 cm were the only independent prognostic factors of better posttreatment survival identified.

Data Availability
No data used to support the findings of the study.

Ethical Approval
This study was performed in accordance with the Declaration of Helsinki (1975) and approved by the Institutional Review Board of 108 Military Central Hospital (O-62720149).

Consent
As this study was retrospective in nature, written consent from the patients was not required.

Conflicts of Interest
The article has no potential conflicts of interest.