The most widely used staging system for hepatocellular carcinoma (HCC) is the Barcelona Liver Clinic Cancer (BCLC) system, which considers tumor burden, performance status, and liver function. Tumor burden is assessed with cross sectional imaging of the abdomen and chest, controversy surrounds the routine use of bone scintigraphy (BS) for detecting extrahepatic metastases. This study evaluated the role of BS in staging HCC in Mexican patients.
Patients and MethodsRetrospective cross-sectional study of all adults with HCC at a Mexican referral center from 2000 to 2018. Staging included abdominal computed tomography (CT) or magnetic resonance imaging, chest CT, and BS. The main outcome was the impact of BS on staging and/or therapy plans.
ResultsAmong 238 patients, 2 with fibrolamellar variant and 44 with incomplete data were excluded. Median age was 66 years, 84 % had cirrhosis, and the predominant etiology was hepatitis C virus (43 %). BCLC stages were distributed as follows: A (14 %), B (7 %), C (68 %), and D (11 %). Extrahepatic disease was present in 18 %; only 8 % patients had a positive BS. Among the positive cases, 4 were true positives, but they did not alter staging or therapy plans.
ConclusionsRoutine BS in HCC staging demonstrated low yield, with a notable rate of false positives. Considering the implications of extrahepatic disease, BS may be justified for liver transplant candidates outside conventional criteria. Our study highlights the limited role of BS in early-stage HCC and advocates for a more selective utilization.
Hepatocellular carcinoma (HCC) is the most common type of liver cancer, and it is the third leading cause of cancer-related death. Its global incidence is decreasing, mainly in Asia, but it is on the rise in Latin America [1]. Diagnosis is usually based on imaging, without the need for a biopsy, and staging is performed based on the Barcelona Liver Clinic Cancer (BCLC) algorithm, which considers three aspects: liver synthetic function, performance status, and tumor burden [2]. The tumor burden includes the size and number of lesions, as well as the presence of vascular invasion and extrahepatic spread. The most common sites of extrahepatic metastases are the lungs, regional lymph nodes, and bones [3,4]. Therefore, staging is based on cross sectional imaging of the thorax and abdomen. Regarding bone scintigraphy (BS), the most recent Mexican Consensus on Hepatocellular Carcinoma recommends its routine use for staging [5], whereas the European Society of Medical Oncology [6] and the National Comprehensive Cancer Network [7] recommend against its routine use (i.e. reserve it for patients with skeletal symptoms), while the American Association for the Study of the Liver [8] and the European Association for the Study of the Liver [9] make no specific recommendation. Therefore, the aim of this study was to assess the role of BS in the staging of HCC in Mexican patients.
2Patients and methodsThis was a cross-sectional study that comprised all adult patients with a new diagnosis of HCC assessed at a referral center in Mexico between the years 2000 and 2018. We excluded patients without BS or with incomplete data that would preclude adequate staging. For the purpose of this study, HCC diagnosis was based on imaging criteria on multiphase computed tomography (CT) or magnetic resonance image in patients with cirrhosis; a liver biopsy was required in cases with atypical imaging features and in all patients without cirrhosis. Imaging-based diagnosis of HCC in patients with cirrhosis was determined based on the European Association for the Study of the Liver guidelines [9]. A liver lesion was considered HCC if it exhibited arterial phase hyperenhancement and washout in the portal and/or delayed phases. Cirrhosis was diagnosed using one or more of the following criteria: histological evidence, radiological findings (ultrasound or cross-sectional imaging showing a lobulated liver and/or unequivocal signs of portal hypertension), or transient elastography with a liver stiffness measurement greater than 14 kPa.
Relevant data was gathered from each patient´s medical record. At our hospital, after diagnosis, all HCC cases are staged with a cross sectional study of the abdomen (either CT or magnetic resonance image), a chest CT, and a BS. The main outcome of the study was the proportion of cases in which the result of the BS modified the stage and/or therapy plan of a patient.
2.1Bone scintigraphyBSs were performed 3 h after intravenous injection of 744 MBq of Tc-99 m methyl diphosphonate. Anterior and posterior whole-body images were acquired. In addition, single-photon emission computed tomography (SPECT) with a dual-head gamma camera (Siemens, Symbia T2 SPECT/CT) was acquired on a case-by-case basis in some body regions when there was diagnostic uncertainty. Low-dose CT imaging was used when relevant, in a targeted fashion on review of BS findings. All studies at our center are interpreted by experienced nuclear medicine physicians. For the purpose of this study, all BSs that showed abnormal findings were reviewed by an experienced nuclear radiologist (EIA), who adjudicated the findings as compatible or not with bone metastasis based on the review of the BS and available CT, SPECT, magnetic resonance imaging, plain radiographs, and also based on the behavior of the lesion over time. Positive BS results, referring to any abnormal uptake that was interpreted in the original report as compatible or suspicious for bone metastasis, were categorized as true-positives when confirmed to be a bone metastasis, or false-positives when the abnormal uptake was due to another reason. This adjudication was done by the nuclear radiologist based on the characteristics of the abnormal uptake, the presence or absence of lesions interpreted as arthritis, osteophytes, or benign compression fractures, and/or the clinical course of the abnormal uptake areas over time. A lesion with a central cold defect was considered malignant.
2.2Sample size and statistical analysesContinuous variables were reported with medians and interquartile ranges, and categorical variables with absolute and relative values. We estimated a sample size of 125 patients to detect a change in staging of 7 % of cases (i.e., proportion of cases migrating from BCLC stage A or B to C), considering a type 1 error of 0.05, and a type 2 error of 0.20 [10].
2.3Ethical statementsThe study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki as reflected in a priori approval by the Ethics Committee of hour hospital (GAS-4137-22-23-1). The need for informed consent was waived due to its observational and retrospective nature.
3ResultsWe reviewed the chart of 238 patients with diagnosis of HCC; we excluded 2 patients who had a primary diagnosis of fibrolamellar variant and 44 patients because of incomplete data (33 without BS, 7 cases where we could not retrieve their imaging studies, 2 with incomplete laboratory results to assess liver function, and 2 due to incomplete clinical/demographic data).
Median age of the population was 66 years (IQR 59-71), 113 were male (59 %), 161 (84 %) had cirrhosis, and the most common etiology of liver disease was hepatitis C virus infection in 70 (43 %) patients. Regarding the stage of HCC, 26 (14 %), 14 (7 %), 131 (68 %), and 21 (11 %) were BCLC A, B, C, and D, respectively. The rest of the general characteristics of the patients can be found in Table 1.
General characteristics of patients (n = 192).
IQR: Interquartile range. MASLD: Metabolic dysfunction-associated steatotic liver disease. MELD: Model for end-stage liver disease.
Thirty-five patients (18 %) had extrahepatic disease. Overall, only 15 (8 %) patients had a positive BS, and of these, only four were true positives for bone metastasis. In two of these cases, the BS showed ribcage lesions that had been missed on cross-sectional imaging, and in the other two, the BS showed lesions in bones that are not routinely visualized when staging HCC (i.e., humerus, femur, skull). In these four cases, the presence of bone metastasis did not change the stage of the disease since the patients were already classified as BCLC stage C due to the presence of extrahepatic non-bone metastases. A false positive BS led to a bone biopsy that ruled out metastasis. Table 2 describes each of the cases with a positive BS.
Patients with bone scintigraphy with abnormal findings.
BCLC: Barcelona Clinic Liver Cancer. BS: Bone scintigraphy. CT: Computed tomography.
The most common HCC staging system is the BCLC, which considers tumor burden, performance status, and hepatic function [2]. The recommended workup for staging includes imaging of the chest and abdomen, but there is inconsistency amongst the international societies regarding the utility of the routine use of BS, as bone metastases are described as the second or third most common site of extrahepatic disease after the lung and lymph nodes [3,4]. On its most recent version of the HCC Consensus, the Mexican Association of Hepatology recommends staging of HCC with BS in all patients [5] with early HCCs undergoing a potentially curative treatment. Therefore, the aim of this study was to assess the role of BS in the staging of HCC in Mexican patients. In our study, we found that most positive BS were false positives, and the 4 true positive BSs did not change the stage or the therapy plan, suggesting there is no role for routinely ordering this study. Our results are in line with those of Koneru B, et al., who studied the utility of BS in 117 patients with early-stage HCC being considered for liver transplantation. In their study, none of the BS were positive, but 9 % were considered indeterminate, and recurrent disease after liver transplant was not different between those with normal and indeterminate BSs (5 % versus 7 %, respectively) [11].
The proportion of patients with bone metastases in our study was only 2 %,which is lower than the 4–24 % reported in other studies [12–17], and this can be because we also included patients with early stage disease. Risk factors for developing extrahepatic disease are those representing the tumor burden (i.e. tumor size, number of nodules, lymph node involvement, intrahepatic metastases, extrahepatic non-bone metastases, and presence of vascular invasion) [4], so it is very unlikely to have bone metastases in early HCC [11,17]. For example, Kutaiba et al. reported 0 % of true-positive BSs in 186 patients awaiting liver transplant [18], and Witjes et al. found bone metastases in only 2 % of 137 patients prior to HCC resection [19]. Therefore, BS should be better reserved for patients outside conservative transplant criteria (e.g. Milan) or for patients that are being considered for downstaging, as a positive BS would have a significant impact in the therapy plan of these patients. Symptom guided investigations are also appropriate, as suggested by case reports of patients with HCC and with atypical initial presentations attributed to bone metastases [20–23], and by series showing that almost all patients with bone disease will be symptomatic [4,24].
In addition to the relatively low frequency of bone metastases, the performance of BS in the specific case of HCC patients is not optimal. Sensitivity is hampered in HCC because most bone metastases are osteolytic in nature [15,25]. In one multicenter study of 211 patients with HCC, 82 % of bone metastases were osteolytic, and the rest were distributed between osteoblastic and mixed [26]. Poor specificity was also evident in our study, in which most positive studies were attributed to non-metastatic bone disease, mainly traumatic and degenerative in nature. Poor specificity may result in over-staging and over-diagnosing, which may result in harm, such as not offering a liver transplant to a patient who otherwise would be an acceptable candidate. Likewise, positive BSs may lead to unnecessary invasive diagnostic procedures and risks, like in our series, in which a patient underwent a bone biopsy. The specificity can be improved by pairing it with a SPECT/CT [27], which can better characterize uptake areas in the BS. However, since bone metastasis is a relatively infrequent event in patients with HCC, and as shown in our study, they usually happen in patients with advanced tumors, there does not seem to be a role for proactively searching for bone metastases in the absence of symptoms. Moreover, as suggested by other authors, routine BS is not cost effective [11,19], and adding SPECT/CT in all cases would increase healthcare costs. Finally, BS interpretation is subject to interobserver variability, adding further uncertainty when using it to stage patients [28].
Our study does have limitations. Due to its retrospective and cross-sectional nature, we were unable to assess if patients with bone metastases were symptomatic, and/or if the findings of the BSs led to any specific interventions, such as palliative radiotherapy, which would further support ordering BSs based on clinical suspicion. In addition, we could not determine the reasons why 33 patients were missing BS results. One of the major limitations is that all BSs were reviewed by a single nuclear radiologist, and as mentioned, interpretation of BSs is subject to interobserver variability. In the particular case of bone metastasis, kappa coefficients are described between 0.53 and 0.88 [29]. However, because the objective of our study was to assess the role of BS in staging, it may be that interobserver variability was not as significant in our study. The nuclear radiologist was able to interpret the findings in light of other available imaging studies and the progression of lesions over time so that any abnormal finding could be retrospectively adjudicated as a bone metastasis.
5ConclusionsIn conclusion, the yield of routinely doing BS as part of the staging of patients with HCC is low and can lead to a significant rate of false positive results. Therefore, we favor a more customized approach, ordering BS on a case-by-case basis based on clinical suspicion. Based on the implications of having extrahepatic disease, BS can also be considered in patients being assessed for liver transplants when outside of conventional criteria. A prospective study focused on this group of patients, intentionally investigating any abnormal findings, would provide important information.
Author contributionsFRA was involved in data curation, methodology, project administration, and investigation. BZMA, RVM, EKO, and LSTM were involved in data curation, writing (original draft), and investigation. GCN, EIA, JAGT were involved in conceptualization, writing (review, editing), and supervision. CMV was involved in conceptualization, writing (review, editing), formal analysis, and supervision and is responsible for the integrity of the work as a whole.
Data availabilityData supporting the findings of this study are available from the corresponding author upon request.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
