Conventional transabdominal US is routinely used as the first-line imaging modality for the detection of liver metastases, and it also plays a substantial role in the intra-procedural localization of lesions during biopsy and therapeutic interventions such as thermal ablation, providing real-time imaging with no ionizing radiation to the operator.

The reported sensitivity of US is much lower than that of CT or magnetic resonance imaging(MRI), ranging from 40 to 77 %, and can drop to 20 % for small lesions (<1 cm) or for lesions that are isoechoic compared with the surrounding liver parenchyma [14, 15]. Thus, detection of hepatic lesions with US frequently initiates a request for an imaging study with greater accuracy, such as CT or MRI.

The development of new ultrasound contrast agents and sonographic techniques has considerably improved the US detection of hepatic metastases and the evaluation of response to therapeutic procedures. Contrast-enhanced ultrasound (CEUS) has improved the sensitivity for detecting LM to 80–90 %, which is not statistically different than the diagnostic performance of CT, and is especially useful for LM smaller than 10 mm [14]. CEUS alone was reported sufficient to classify 89.0 % of focal liver lesions as benign or malignant, which served as a one-stop diagnostic test for 80.8 % of patients, reducing the need for CT and MRI scans, and provided savings in terms of radiation exposure, time, and money [16].

In clinical practice, CT is the mainstay of oncological imaging because of its excellent availability, good patient tolerance, and reproducibility. As metastatic hepatic lesions are usually hypovascular, the optimal CT strategy is helical scanning during the portal venous phase of enhancement. This technique improves lesion identification by increasing the attenuation of normal liver tissue, and occasionally rim enhancement of a hypoattenuating metastasis can be seen [17].

The advent of multidetector row computed tomography (MDCT) effectively replaced computed tomography during arterial portography for diagnosis of hepatic metastases. MDCT is highly accurate for evaluating staging, especially for the detection of lymph node enlargement and distant metastases. The accuracy of MDCT for overall M staging is 85.6 %, and for identifying LM the sensitivity is 80.0 % [18]. CT is a standard test for the detection of the LM; however, the reported sensitivity for small metastases (<1.5 cm) is low [19]. A recent comparison between total liver-volume perfusion CT (CTP) and four-phase CT revealed that CTP increased the sensitivity for the detection of LM to 89.2 % from 78.4 % (P = 0.046) and specificity to 82.6 % from 78.3 % (P = 0.074), indicating that CTP is a noninvasive, quantitative, and feasible technique [20].

MRI is considered to be the optimal diagnostic modality for evaluation of suspected LM, with a reported sensitivity of 80–100% and specificity up to 97 % [21]. The diagnostic performance of MRI for small lesions (<1.0 cm) has been shown to be superior to helical CT [22]. Modern comprehensive liver imaging protocols generally use a variety of sequences to optimize lesion detection and characterization. Diffusion-weighted (DW)-MRI has been found to be sensitive for the detection of small liver lesions (<1 cm) that mimic small intrahepatic vessels; however, as there is an overlap of apparent diffusion coefficient (ADC) values between different types of lesions, the predominant limitation of the DW-echo planar imaging (EPI) sequence is the differentiation between benign and malignant lesions [23]. Contrast-enhanced MRI can demonstrate tissue-specific physiological information, thereby facilitating liver lesion characterization, which may facilitate the accurate diagnostic workup of patients in order to avoid invasive procedures for lesion characterization, such as biopsy. Gadolinium-enhanced T2-weighted images and superparamagnetic iron oxide (SPIO)-enhanced T2-weighted MRI have been reported to have the ability to differentiate primary malignant, metastatic, and benign lesions in the liver [24]. However, with the increasing awareness of the potential risks of contrast material administration in patients with severe renal failure, there is a strong clinical need for methods to diagnose LM without exogenous contrast materials. Nasu et al. [25] retrospectively compared the accuracy of respiratory-triggered DW-EPI in combination with unenhanced T1- and T2-weighted imaging versus SPIO-enhanced imaging and found that there was no significant difference of sensitivity and specificity between the two imaging methods. DW-MRI combined with unenhanced T1- and T2-weighted imaging is a reasonable alternative to contrast-enhanced MRI for the detection of LM [25].

18F-FDG PET and FDG-PET with CT (PET/CT) have the added advantage over MRI and CT of providing not only anatomic but also functional information. The major advantages of PET are the ability to detect extrahepatic disease and to evaluate therapeutic response early. One series found that FDG-PET had a sensitivity and specificity of 85 and 74 %, respectively, for the detection of LM originating from gastric cancer [26]. A recent study demonstrated that the detection rate of liver lesions was significantly lower for PET/CT than for gadolinium-ethoxybenzyl-diethylenetriamine pentaacetic acid-enhanced MRI (64 % vs. 85 %, respectively; P = 0.002), which is especially relevant in small LM (≤1 cm in diameter) [27].

Luteinizing hormone-releasing hormone (LHRH) agonists have become the standard of care in castration therapy because these agents have the potential of reversibility and enable the use of intermittent ADT, avoid the physical and psychological discomfort associated with orchiectomy, have a lower risk of cardiotoxicity than observed with diethylstilbestrol, and result in equivalent oncologic efficacy [28, 29].