We assessed the relationship between advanced age and the outcome of multidisciplinary treatment for CRLMs. The study revealed that cancer-related survival in patients ≥ 75 years was significantly impaired, which may have been caused by the lower rate of repeat hepatectomy for liver recurrence in the older patients. However, liver resection can provide an acceptable prognosis with short-term outcomes comparable to those of the younger patients.
In the current study, the 5-year DSS rate in the elderly group (44.2%) was significantly lower than that of the younger patients. Multivariate analysis of the study population revealed that age ≥ 75 years was an independent predictor for impaired DSS. There are many reports assessing the prognostic benefit of hepatectomy in the elderly[5, 7, 9, 11, 14, 16, 17]. However, the mixed reporting of overall and cancer-specific survival has made it difficult to interpret the long-term outcome. To investigate the differences in chronological age, DSS needs to be assessed because OS is impaired in the elderly as a result of the more limited life expectancy than in younger patients. Brudvik et al. well demonstrated a significant difference between OS and DSS in the elderly (80–89 years of age) and the prognostic benefit of hepatectomy by showing that the gap between the 5-year survival of the age-matched national population (66.3%) and 5-year DSS rate (43.1%) decreases compared with the 5-year OS rate (32.5%) after hepatectomy[17].
Analysis of recurrence after initial hepatectomy revealed that repeat hepatectomy was performed less in the elderly group with liver recurrence, although the recurrence pattern was not different between the groups. This result indicated that in clinical practice, aggressive treatment for recurrence was reserved, even when indication criteria for repeat hepatectomy were met, regardless of the patient’s age. The major strength of this study was the detailed analysis of treatment for recurrence after initial hepatectomy, which revealed the cause of the poorer disease-specific prognosis in the elderly having CRLMs. Repeat hepatectomy is one of the essential treatments for CRLMs, because the biology of colorectal cancer is unique; the recurrence after the initial surgery is not directly associated with cancer-related death. Oba et al. demonstrated that the time to development of unresectable recurrence after initial surgery is a better surrogates prognosis than RFS in patients undergoing surgery for CRLMs[23]. The current study revealed that repeat hepatectomy was safely performed without mortality even in the elderly, which is consistent with a previous report showing no mortality in 114 patients ≥ 70 years having repeat resection after initial hepatectomy[7]. The results of the current study confirm that aggressive repeat resection is a feasible option to improve survival in well-selected older patients having recurrence after initial hepatectomy. The next step is to elucidate the indication process for repeat hepatectomy in the elderly so as to optimize patient selection for the aggressive treatment.
The prevalence of adjuvant chemotherapy after initial hepatectomy was lower in the elderly group than in the younger groups. This result may be another cause of poorer DSS in the elderly in this study. Adam et al. demonstrated that no adjuvant chemotherapy after initial hepatectomy was an independent predictor of reduced OS in the patients ≥ 70 years undergoing hepatectomy for CRLMs[7]. However, their analysis of OS rather than DSS has made it difficult to interpret the result because senile weakness intolerant to adjuvant chemotherapy may have been directly associated with cancer-unrelated death in their assessment. Although administration of adjuvant chemotherapy after hepatectomy was not selected as a prognostic factor of DSS in univariate and multivariate analysis in the current study, a recent randomized trial demonstrated that adjuvant chemotherapy improves RFS in patients undergoing hepatectomy for CRLMs[27]. Additional studies are needed to investigate the prognostic impact of adjuvant chemotherapy, especially in older patients for whom the balance between therapeutic effect and toxicity is of great importance.
Initial and repeat hepatectomies were safely performed in the elderly group without mortality in the current study. No differences were found in other short-term outcomes, such as the prevalence of major complications and the length of hospital stay after initial hepatectomy, even when surgical procedures were similar among the groups. We assume that the favorable short-term outcomes in the elderly were attributable to the low rate of major hepatectomy (19.6%) in this study, considering the results of previous reports. Although there is some discrepancy in the definition of “the elder” and “mortality,” previous large series of population-based or multicenter studies demonstrated the rates of major hepatectomy as 37.5% to 56% in the elderly. Consequently, the mortality rate was reportedly as high as 3.8% to 8%[7, 12, 14]. The parenchymal-sparing approach is now accepted as the standard procedure for resection of CRLMs to achieve better short- and long-term outcomes[28-30]. Greater concern should be taken to choose less invasive parenchymal-sparing hepatectomy in the elderly who are physically weak because of senile decay. Referral to a specialist hepatobiliary surgery team is favorable to avoid major hepatectomy because parenchymal-sparing hepatectomy for tumors in difficult locations is technically demanding[31].
The limitations of this study include its retrospective nature and the small number of patients in a single-center experience. As mentioned above, the selection process for repeat hepatectomy was not clear because of the limited data. However, the detailed analysis of recurrence pattern and the treatment for recurrence would not have been possible using population-based data[12, 14]. In addition, although population-based analyses are said to better describe the outcomes achieved in routine practice[14], the trend in centralization of high-risk surgery is associated with improved short- as well as long-term outcomes[32-42]. Considering much better outcomes reported from high-volume liver centers[20, 43-45], the results demonstrated in the current study may reflect ideal practice in the near future when centralization is optimized for older patients with CRLMs undergoing hepatectomy.
In conclusion, in patients ≥ 75 years undergoing hepatectomy for CRLMs, cancer-related survival was significantly impaired, which may have been caused by the lower rate of repeat hepatectomy for recurrence in this population. However, liver resection can provide an acceptable prognosis with short-term outcomes comparable to those of the younger patients.