From January 2001 to December 2010, 103 consecutive patients were diagnosed as having lung metastasis at the first time of HCC diagnosis in the 5th Department of Hepatobiliary Surgery, Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China. To evaluate the efficacy of anti-tumor treatment modality for patients with HCC and concomitant isolated lung metastasis, we excluded those patients who showed evidence of severe organ failure (renal, respiratory, or cardiologic problems), poor liver function (Child-Pugh class C) that could affect survival, or co-existing metastasis from other sites. Therefore, 76 of the 103 patients (59 males and 17 females) were recruited for the study. The last follow-up date was January 30, 2011. This study protocol was approved by the Institutional Review Board of the Eastern Hepatobiliary Hospital.

We used a prospectively maintained database and conducted a retrospective study among these patients. We focused on the prognosis and prognostic factors affecting the survival of HCC patients presenting with lung metastasis. To investigate whether local regional or systemic therapy could affect survival, we stratified these 76 patients into three groups: untreated group (n = 22), single treatment group (n = 19), and combined treatment group (n = 35). All of the patients in the untreated group refused any anti-tumor invasive treatment other than conservative treatment, including the support of liver function, after being informed of the expenses and the possible side effects of any anti-tumor invasive treatment. The single treatment method for these patients included hepatectomy, TACE, radiotherapy, chemotherapeutics, and oral sorafenib, while combined treatment modalities were defined as more than one of the above treatment methods or pulmonary metastasectomy.

Laboratory blood tests including hepatitis B virus (HBV) markers, anti-hepatitis C virus, serum α-fetoprotein (AFP), carcinoembryonic antigen, platelet, serum albumin, serum total bilirubin, alanine transaminase, aspartate aminotransferase, and prothrombin time were performed.

The diagnosis of HCC was based on concordance between two imaging examinations [ultrasonography, computed tomography (CT), or magnetic resonance imaging (MRI)] showing arterial hypervascularity in a focal lesion ≥ 2 cm or with the combined criteria of an imaging examination and a serum AFP level greater than 400 ng/mL, according to the criteria of the Conference of the European Association for Study of the Liver[5]. In the present study, the diagnosis of HCC was also confirmed by pathohistology in 20 patients.

Lung metastases were diagnosed in all the 76 patients using imaging techniques, including chest CT (n = 62), positron emission tomography-CT (n = 11), and chest MRI (n = 3). In addition, one patient was further confirmed by pathohistology after a biopsy, and four patients after pulmonary metastasectomy.

Continuous data were expressed as mean ± SD or median (range). Categorical variables were compared by the χ² test or Fisher exact test, and continuous variables were compared by the Student t test or one-way analysis of variance. The survival rate was calculated using the Kaplan-Meier method, and the log-rank test was used to compare survival rates among three groups. Cox’s proportional hazards model was used for multivariate analysis. All statistical analysis in this study were done using software package SPSS11.0 (SPSS Inc., Chicago, IL). A P value < 0.05 was defined to be statistically significant.

The baseline characteristics of these 76 patients with advanced HCC initially presenting with lung metastasis are summarized in Table 1. The mean age of patients was 52.4 ± 10.7 years. The most common etiology of the liver disease was HBV infection, which occurred in 63 patients (82.9%). There was no statistical difference in the baseline characteristics among the untreated, single treatment and combined treatment groups, including age, etiology of the liver disease, the presence of cirrhosis or ascites, AFP level, Eastern Cooperative Oncology Group (ECOG) scale, Child-Pugh score, and liver function.

For intrahepatic tumor status, there was no significant difference in tumor number, tumor location, and the probability of portal vein tumor thrombus or hepatic vein tumor thrombus among the three treatment groups, except for maximum tumor diameter (P = 0.007) (Table 2). As such, there is also no significant difference in number, location, and maximum diameter of lung metastatic nodules among these three groups (Table 2).

Among all the 76 patients, 22 patients did not receive any anti-tumor treatment but conservative treatment. We divided the remaining 54 patients into two groups according to their treatment schemes. Nineteen patients (25.0%) received single treatment modality (locoregional or systemic therapy), including hepatectomy in 4 patients, TACE in 6, radiotherapy for intrahepatic tumor and lung metastatic nodules in 5 (including γ knife radiosurgery in 3 and X-ray radiotherapy in 2), and oral sorafenib in 4. Thirty-five patients (46.1%) received combined treatment modalities (Table 3). The common treatment modality in the combined treatment group was radiotherapy (30/35, 85.7%), followed by hepatectomy (23/35, 65.7%), TACE (18/35, 51.4%), and oral sorafenib therapy (11/35, 31.4%). In the combined treatment group, radiotherapy (γ knife radiosurgery in 24 and X-ray radiotherapy in 4) was mainly used for lung metastatic nodules (27/28) rather than intrahepatic tumor nodules (8/28). In addition, pulmonary metastasectomies were carried out in 4 patients who underwent hepatectomy during the same operation.

After a median follow-up of 8.7 ± 0.6 mo (range, 1.1-68.7 mo), 59 patients died and 17 patients remained alive. The causes of mortality were disease progression, including hepatic failure in 31 patients (52.5%); cachexia in 12 (20.3%); upper gastrointestinal bleeding in four (6.8%); pneumonia in one (1.7%); and pulmonary thromboembolism in one (1.7%). The cause of mortality of 10 patients was not confirmed. Among 17 surviving patients, only 4 patients were disease-free, including 3 patients who underwent hepatectomy and pulmonary metastasectomy, and one patient who received hepatectomy and subsequent γ-knife radiosurgery. Among 13 patients who were not disease-free but alive, 6 patients still took medicine of sorafenib, and achieved complete or partial response on radiographic examination (Figure 1A and B). The 6-mo, 1-year and 2-year cumulative survival rates of all these patients were 64.5%, 40.1% and 18.5%, respectively.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Overall survival. A: Computed tomography (CT) scan showing multiple metastatic nodules in both lung lobes before anti-tumor treatment in a 54-year-old male patient with lung metastasis due to hepatocellular carcinoma; B: CT scan showing the disappearance of previous lung metastatic nodules in 5 mo after γ-knife radiosurgery and oral sorafenib.

The median survival time of combined treatment group was 18.6 ± 3.9 mo, which was longer than that of single treatment group (6.3 ± 2.5 mo) and untreatment group (4.1 ± 0.3 mo), and there was significant statistical differences among these groups (log-rank test, P < 0.001) (Figure 2). The 6-mo cumulative survival rate of combined treatment, single treatment and untreated group was 88.6%, 57.9% and 31.8%, respectively; and the 1-year survival rate of the three groups was 68.5%, 21.1% and 8.0%, respectively.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Comparison of survival among three treatment groups. Survival curves of untreated group (n = 22), single treatment group (n = 19), and combined treatment group (n = 35).

In investigation of prognostic factors of survival of HCC presenting with lung metastasis at initial diagnosis, univariate analysis showed that ECOG score, Child-Pugh score, ascites, AFP level, maximum intrahepatic tumor diameter, the absence or presence of portal vein tumor thrombus, the absence or presence of hepatic vein tumor thrombus, lung metastatic tumor number, and treatment modality had prognostic significance (Table 4).

In the Cox proportional hazard model, Child-Pugh score, the absence or presence of portal vein tumor thrombus at the initial presentation, and treatment modality were three independent prognostic factors that affected the survival of HCC patients that presented with lung metastasis at initial diagnosis (Table 4).

The prognosis and treatment outcomes of advanced hepatocellular carcinoma (HCC) presenting with lung metastasis remain poorly evaluated.

There are various treatment modalities for advanced HCC, including surgical resection, transcatheter arterial chemoembolization, radiotherapy, chemotherapeutics, and administration of molecular targeted therapeutic drugs. This study investigated the prognosis, treatment outcomes, and independent prognostic factors affecting the survival of a series of HCC patients presenting with lung metastasis at initial diagnosis.

In this study, the survival times from date of diagnosis until death or last visit between untreated group, single treatment group, and combined treatment group were significantly different, and patients underwent specific combined treatment modalities had longer survival than those treated with single method or those without anti-tumor treatment. Multivariate analysis revealed that Child-Pugh score, the absence or presence of portal vein tumor thrombus, and treatment modality were three independent prognostic factors affecting survival of patients with advanced HCC and concomitant lung metastasis.

As one type of advanced HCC, HCC presenting with lung metastasis is not unusual, and its prognosis is very poor. The findings in this study may contribute to its prognosis, and combined treatment modalities tend to result in a better survival for patients with advanced HCC initially presenting with lung metastasis.

The authors firstly investigated that combined treatment modalities tended to yield better survival prolongation compared with conservative treatment or single treatment modality for HCC patients initially presenting with lung metastasis, which may be contribute to its prognosis.