Contrast-enhanced ultrasound (CEUS) - A new tool for evaluating blood supply in primary peripheral lung cancer.

OBJECTIVES: To explore the feasibility of contrast-enhanced ultrasound (CEUS) as a new tool for characterizing vascularization of primary peripheral lung cancer. METHODS: 315 consecutive patients with definite primary peripheral lung cancers underwent CEUS examination from November 2016 to March 2022. CEUS parameters including time to enhancement (TE), time to peak (TP), time to wash-out (TW), distribution of vessels (DV), extent of enhancement (EE) and homogeneity of enhancement (HE) were obtained. RESULTS: The lesions were grouped on the basis of TE which reflects tumor vascularization: early enhancement (pulmonary arterial vascularization) (n = 91) and delayed enhancement group (bronchial arterial vascularization) (n = 224). Overall, lung tumors commonly (71.1%) manifested a delayed enhancement which indicating blood supply originated from bronchial arteries, while an early enhancement was present in less than a third of the cases. Tumors with bronchial vascularization tended to show a delayed, reduced and heterogeneous enhancement. Correspondingly, it is characterized by a shorter TE, marked EE and a relatively infrequent occurrence of necrosis in tumors with pulmonary vascularization. CONCLUSIONS: Providing micro-perfusion information, CEUS is a potentially imaging tool for evaluating blood supply in primary peripheral lung cancer.

Keywords: Contrast-enhanced ultrasound; tumor vascularization; lung cancers

Tumor growth is a complex process, the mechanism of which is unclear. Although angiogenesis has been evidently demonstrated a vital factor in tumor sustainable development, how it manifests in the lung become complicated by the fact that the organ has double circulations (the pulmonary and systemic bronchial circulation) [[1]]. Therefore, studying angiogenesis and blood supply, may be essential for tissue characterization and treatment selection of pulmonary carcinoma [[2]]. Clinically, computed tomography (CT) scan with dynamic contrast enhancement remains workhorse for identifying pulmonary carcinomas [[3]]. Over the past decade, efforts have been made to assess the angiogenesis of lung cancer by using chest computed perfusion (CTP) imaging, which is non-real-time, expensive and especially has high radiation burden [[4]].

Ultrasonography (US) of the lung diseases has been considered experimental for many years. Sonographic examination of periphery pulmonary lesions by ultrasound is possible when the difference in impedance between focal tissue and total reflecting lung tissue is reduced. For decades, the application of contrast-enhanced ultrasound (CEUS) in lung has gradually drawn great attentions since it offers detailed information on tumor vascularity [[5], [7], [9]]. As an appositional imaging tool, CEUS is –up to now –not the primary diagnostic method. There is also no general recommendation by European Federation for Ultrasound in Medicine and Biology (EFSUMB) for CEUS in pulmonary tumors [[10]]. However, angiogenesis as previously noted is a crucial factor in tumor growth and the recognition of where neo-angiogenesis derived from is of great diagnostic and therapeutic interest [[11]].

Based on the hypothesis that CEUS features might be different depending on blood supply, a few exploratory researches have been reported [[6], [13]]. Therefore, the purpose of this work was to assess the feasibility of CEUS as a new tool for characterizing blood supply of primary periphery lung cancer in a larger series data by using the dual "in vivo" references of gas-containing pulmonary tissues and chest wall.

From November 2016 to March 2022, n = 511 consecutive patients (357 men and 154 women; mean age, 61.06±11.24 years; range, 18–84 years) with suspected primary peripheral pulmonary carcinoma checked by contrast-enhanced computed tomography (CECT) were underwent ultrasound-guided percutaneous biopsy. B-mode US as well as CEUS examinations were performed before biopsy. 196 patients were excluded because histology results revealed lesions other than primary lung cancer (pneumonia = 111, tuberculosis = 78, pulmonary embolism = 1, atelectasis = 3, fibroblasts hyperplasia = 1, metastasis = 2). Finally, 315 patients were included in the study. This study was approved by the ethics committee of Lanzhou University Second Hospital (Approval No. 2022A-201) that waived informed consent for the retrospective data analysis of the patients. Additionally, a written consensus was signed by all patients before each examination.

Sonographic examinations were carried out by a Philips iU22 or EPIQ 7 device (Seattle, USA), which is equipped with a C5-1 convex probe operated at a center frequency of 1∼5 MHz. Baseline US features such as echogenicity, maximal size, homogeneity and air bronchogram were investigated. CEUS was achieved by a second-generation contrast agent SonoVue (Milano, Italy). Prior to injection, sulfur hexafluoride suspension was obtained by adding 5 ml 0.9% sodium chloride solution in lyophilized powder followed by thorough shaking. A bolus of 2 ml suspension then was injected into median cubital vein; this immediately followed by a 5 ml saline flush. Dynamic videos were recorded and stored.

In CEUS the following parameters were evaluated: time to enhancement (TE), time to peak (TP), time to wash-out (TW), distribution of vessels (DV) within tumors (tree-like, chaotic, indefinite pattern) [[14]], extent of enhancement (EE: hyper-/iso-enhancement, hypo-enhancement) and homogeneity of enhancement (HE: homogeneous, heterogeneous). With the "in vivo" references of gas-containing pulmonary tissues and chest wall, lesions with early enhancement (contemporaneously or immediately following the enhancement of lung) were considered to have a pulmonary arterial (PA) blood supply; delayed enhancement (at least 2 seconds after the enhancement of lung or at the same time or later than the enhancement of chest wall) was described as bronchial arterial (BA) vascularization [[15]]. ΔTP and ΔTW were defined as the differences between TP and TE, TW and TE, respectively.

All examinations were performed by two authors (T. D and T. L with 3 and 5 years of experience in CEUS, respectively). In case of disagreement between the two, a third reviewer (F. N with more than 18 years of experience in the field) was consulted for final consensus interpretation. The final diagnosis was based on histopathologic findings of ultrasound-guided percutaneous biopsy or surgical specimens.

Statistical analysis was performed via SPSS 23.0 statistical software (IBM Corporation, Armonk, NY). Statistical evaluation of categorical data was based on the chi-square test. The two-independent sample t-test was applied to compare the differences in the measurement data. P < 0.05 was considered statistically significant.

The 315 patients consisted of 241 men and 74 women, with the mean age of 63.70±9.12 years (range, 30–84 years). The final diagnosis revealed the study cohort included 155 adenocarcinomas, 78 squamous cell carcinomas, 40 small cell lung carcinomas, 4 large cell neuroendocrine carcinomas, 1 carcinoid tumor, 6 large cell carcinomas, 6 adenosquamous carcinomas, 13 sarcomatoid carcinomas, 2 synovial sarcomas, 1 MALT lymphoma, 2 malignant solitary fibrous tumors and 7 undifferentiated pulmonary carcinomas. Overall, the blood supply to lung tumors largely originated from bronchial arteries (224, 71.1%) and a pulmonary arterial vascularization was present in 28.9% of the cases. Details of the blood supply for different pathological type are shown in Table 1.

Table 1 Blood supply of primary peripheral lung cancer regarding different pathologies

Also, both groups were stratified by sex, age and smoking. In this study cohort, men were a significant portion of the gender composition. And each respectively, the percentages were 79.9% and 68.1% in the early enhancement group and the delayed enhancement group, the difference was significant (χ 2 = 4.995, P = 0.025). But there were no evidences that age (t = 0.010, P = 0.992) and smoking (χ 2 = 0.394, P = 0.530) have influences on the blood supply of lung cancer.

On average, quantitative time parameters (TE, TP and TW) in the early enhancement group were faster than the delayed enhancement group (all P < 0.001). There was no statistical evidence that ΔTP is different between the two groups (P = 0.285). Neither significant difference was found in ΔTW (P = 0.257). Generally, DV within tumors commonly manifested a chaotic pattern (174/315, 55.2%). The bubbles flow with a chaotic or indefinite pattern in a large fraction of delayed enhancement cases (219/224, 97.8%) while tree-like filling was predominantly seen in the early enhancement group (22/27, 81.5%). The early enhanced lesions showed an EE in the sense of hyper-/iso-enhanced (63/91, 69.2%) to hypoenhanced (28/91, 30.8%). In the delayed enhancement group, relative percentages of hyper-/iso-enhancement and hypoenhancement were 45.1% and 54.9%, respectively. While homogenous enhancement was observed in approximately two thirds of the lesions with PA supply (60/91, 65.9%), necrosis occurred mainly in the delayed enhancement group (122/153, 79.7%) (Figs. 1, 2). For a detailed overview of the CEUS features regarding type of blood supply, see Table 2.

Graph: Fig. 1 58-year-old male with adenocarcinoma. a. CT demonstrates a consolidation in right lower lobe. b. US shows a hypoechoic, triangular lesion. c. twelve seconds after injection of contrast agent, the lesion (arrowhead) enhances simultaneously with adjacent aerated lung tissue (arrow), that is to say an early enhancement. d. tree-like vessels are clearly visible within the tumor. e. CEUS shows a homogenous, marked enhancement at peak. f. twenty-one seconds, chest wall and liver (arrowhead) that supplied by systemic circulation start enhancing.

Graph: Fig. 2 69-year-old male with adenocarcinoma. a. CT shows a lesion in right upper lobe. b. baseline US demonstrates a hypoechoic, oval mass. c. the lung tissue (arrowhead) adjacent the lesion enhances first at six seconds after injection of contrast medium. d. twelve seconds, the tumor (arrowhead) performs simultaneous enhancement with the chest wall (arrow). e. vessels within the tumor manifests a chaotic pattern (arrowhead). f. a heterogeneous enhancement with central anechoic areas (*) due to necrosis is seen at enhancement peak (twenty-one seconds).

Table 2 CEUS parameters in 315 lung cancers stratified by pulmonary and bronchial vascularization assessment

Abbreviations: TE, time to enhancement; TP, time to peak; TW, time to wash-out; DV, distribution of vessels; EE, extent of enhancement; HE, homogeneity of enhancement.

Lung cancer is the leading cause of cancer-associated death around the globe [[16]]. Similar to that of most other cancers, long-term prognosis for lung cancer is directly bound up with the stage at the visit time. Unfortunately, most cases are confirmed in its advanced stage and thus become inoperable, the five-year survival rate of which is very low [[17]]. If transarterial interventional therapy is contemplated, knowledge about the blood supply of pulmonary tumors may provide guidance for choosing an appropriate approach [[18]].

The blood supply of lung cancer is still controversial, most publications demonstrated that vascularization of lung cancer originates predominantly from systemic circulation with pulmonary arteries involved potentially [[20], [22]]. The advent of CTP imaging enabled the quantification of vascular function, however in lungs, it remains a challenge due to breathing motion and radiation stress [[4], [11]]. Unlike iodine-based contrast media used in CECT, using a blood pool agent, CEUS can provide more perfusion details of the lesion and allow monitoring in real time [[24]]. A variety of different perfusion software for CEUS enables qualitative and quantitative analysis of tumor microvascularization and perfusion [[25]]. Thus, CEUS seems to be an accurate imaging technology to assess the tumor vascularity. Based on above, this work aimed to explore the feasibility of CEUS as a new tool for evaluating blood supply in primary peripheral lung cancer.

With regard to quantitative parameters, the results are consistent with previous researches that TE is a significant indicator to distinguish blood supply [[15], [26]]. But multiple factors can affect and change the standard time window of both pulmonary and systemic bronchial circulations, including intrapleural pressure gradient, the Euler-Liljestrand reflex (hypoxic vasoconstriction), thyroid dysfunction, and so on [[27]]. To better minimize the above impacts, a pulmonary circulation organ (normal lung tissue) and a systemic circulation organ (chest wall) were used as references in each case. TE was classified as early or delayed accordingly. A delayed TE of the lesion indicates bronchial vascularization. In the present study, TP and TW in the BA vascularization group were longer than the PA vascularization group (P < 0.001), but no significant difference was detected in ΔTP and ΔTW. One of the possible reasons is that these two parameters appeared to be affected by TE.

In general, the bronchial circulation is more transitional and has a higher angiogenetic capacity under the pathological circumstances [[29]]. This can explain that vascularization of neoplastic lesions mainly originates from bronchial arteries, characterizing an inhomogeneous, hypoechoic tissue enhancement [[30]]. Our results also lend support to the standpoint. Over 70% of peripheral pulmonary tumors in this series, especially adenocarcinoma, squamous cell carcinoma and large cell carcinoma, was primarily supplied from bronchial arteries. These lesions tended to show a delayed, reduced and heterogeneous enhancement, with a chaotic distribution of tumor vessels.

Generally, because of a blood supply originated from the pulmonary circulation, benign lesions such as inflammatory processes and atelectasis are characterized by a rapid influx of contrast media (shorter TE) with homogeneous, marked enhancement [[5], [27]]. But interestingly, the same manifestations were observed in the nearly three-tenths of the tumors. Histologically, on the one hand, there exist the pronounced inflammatory environment in certain pulmonary tumors such as lymphoma [[31]]; On the other hand, a portion of lung cancers (especially certain adenocarcinoma subtypes) can exploit pre-existing pulmonary vessels to support growth if a suitable vascular bed is available [[20], [32], [34]]. It is hence easy to understand why dendritic distribution of vessels is more often seen in the PA vascularization lesions. Tuberculomas and inflammatory pseudotumors have been very difficult to distinguish from malignant pulmonary tumors to date because of the special pathological processes, resulting in a very low diagnosis rate associated with various imaging techniques [[5], [35], [37]]. Therefore, when we face indications given by CEUS, we also should integrate and consider other clinical information and reach a correct diagnosis. Furthermore, the two vascularization types were approximately owned on a fifty-fifty basis in small cell lung carcinoma. Generally, tumors which grow initially by using ready-made pulmonary capillary network is more aggressive and may become resistant to some anti-angiogenetic therapy [[20]]. In this context, awareness of blood supply in pulmonary tumors would be conductive to select an appropriate treatment and prolong the survival duration.

The present study has several shortcomings. First, the study population was almost entirely composed of the patients with ultrasound-guided percutaneous biopsy, inevitably leading to existence of certain bias. Simultaneously, further classification of specific pathological types (e.g., adenocarcinoma) based on small biopsies is scarcely feasible. This, yet, does not affect the overall results. Besides, due to the limitations of the CEUS technology itself, we were unable to directly and independently measure the weighting of the pulmonary and bronchial vessels in the overall tumor perfusion. It is also a tremendous challenge confronted with CTP imaging.

In conclusion, the feasibility of CEUS as a new tool to evaluate blood supply in primary peripheral lung cancer has been described in the work. It demonstrates two central features: one, the blood supply in primary peripheral lung cancer is influenced by its pathological type, but the bronchial vascularization is usually dominant; the other, lung cancers with different originations of blood supply have different CEUS characteristics.

This study was funded by the National Natural Science Foundation of China (Grant No. 81873898).