PJS affects approximately one in 100,000 individuals and is characterized by black macules in the labial regions, oral cavity, and extremities, as well as multiple gastrointestinal polyps [16, 17]. PJS is attributed to a mutation in the gene encoding serine-threonine kinase 11 tumor suppressor and is autosomal dominant [18, 19]. Accordingly, PJS is also called familial mucocutaneous hyperpigmented gastrointestinal polyposis. Females with PJS may develop gynecological tumors, which include various pathologic types, such as SCTAT, LEGH, G-EAC, ovarian mucinous tumor, and endometrioid adenocarcinoma [6].

The diagnosis of PJS can be established if any one of the following criteria is met: (1) at least two histologically confirmed PJ polyps; (2) at least one histologically confirmed PJ polyp and family history of PJS in close relatives; (3) pigmented moles on skin and mucosa and family history of PJS in close relatives; (4) at least one histologically confirmed PJS polyp and pigmented moles on skin and mucosa [12]. A diagnosis of G-EAC is mainly based on pathological examinations combined with immunohistochemical staining. Notably, PJS and G-EAC frequently cooccur in a single individual; therefore, if PJS or G-EAC is diagnosed or suspected, the possibility of the other should be carefully examined. A pelvic MRI should be conducted in a timely manner. Moreover, a deep cervical biopsy under US guidance, segmented curettage, and cervical conization should also be performed when necessary [20].

For the management of PJS, the surveillance of gastroenterological polyps and the occurrence of cancer is of marked relevance, as no radical therapy is available for curing PJS [12]. Owing to age-dependent features of PJS, the early diagnosis of sizable gastrointestinal polyps and timely surgical interventions can avoid intussusception and anemia in young patients. Regular screening and surgical removal of cancerous lesions at early stages can substantially improve the prognosis of adult patients.

Since no standard therapy has been established for addressing G-EAC, personalized treatment is highly recommended for individual patients. During the early stages of G-EAC, the removal of cervical lesions and metastases, if any, by surgical intervention is critical. Furthermore, chemoradiotherapy should be undertaken postoperatively. For advanced-stage G-EAC, chemoradiotherapy remains the predominant treatment method [20, 21]. Reportedly, the expression of human epidermal growth factor receptor-2 (HER-2) is commonly observed in patients of G-EAC with ovarian metastasis or in advanced stages; hence, targeted therapy using HER-2 monoclonal antibodies, such as trastuzumab, has been deemed a feasible therapeutic option for G-EAC [22].

During early-stage G-EAC, the macroscopic appearance of the cervix is relatively normal, as the lesion is hidden in the middle and upper part of the cervical canal, making it markedly difficult to be sampled during routine screening. Moreover, owing to the mildly heterotypic cytological feature of G-EAC, the positive rate of G-AEC in cervical exfoliative cytology and cervical biopsy examinations is only 32.7% [23]. Therefore, conventional cervical cancer screening methods fail to meet the clinical needs for diagnosing this disease. These factors highlight the importance of medical imaging in diagnosing G-EAC and distinguishing benign and malignant cervical lesions using supplementary methods [7, 24, 25].

We briefly summarized the major features of G-EAC examined by different medical imaging modalities in Table 2. Given the low soft tissue contrast resolution, computed tomography (CT) is infrequently applied to diagnose G-EAC. Under CT imaging, G-EAC presents an enlarged cervix comprising masses of low intensity, most of which are multiple complex cystic masses with smooth edges, and a few are solid masses with blurred edges [7, 8, 26,27,28]. Contrast-enhanced CT can provide more details and even reveal cystic lesions with enhanced contrast, which cannot be detected under ordinary CT examination.

According to certain specialists, MRI may better reflect the histological architecture of the cervical lesion, detect its solid components, and present more detailed features than CT [8, 24, 26, 29]. Based on the pathomorphological and distribution patterns of lesions, the typical MRI patterns of G-EAC were classified into four types including “cosmos”(small cystic components encompassed by bigger cysts), “diffuse growth”(multiple cysts of small size and diffuse distribution), “focal mass-like bulging”(unilaterally distributed numerous microcysts forming a focal bulge that grows outward of the cervical wall), and “solid and cystic”(bilaterally distributed multiple cysts of various sizes and mixed components). The first two types are deemed benign lesions; the third type is precancerous, whereas the last type indicates malignancy [30]. Typically, in most cases, G-EAC exhibits barrel-like multicystic massed with hyper-intense in T2-weighted images, comprising solid components that grow from the cervical glands into the stroma, replacing the original fibromuscular cervical stroma [24]. In certain cases, G-EAC lesions in the T2 image may be purely solid or cystic [23, 28, 31]. In contrast, the postcontrast T1-weighted images exhibit fat saturation, revealing the mildly heterogeneous enhancement of the solid components inside the cervical lesion [24]. Importantly, the solid lesion might be correlated with elevated risks of invasion and metastasis, which, accordingly, is a critical feature for distinguishing the malignancy of the multilobular cystic masses because non-malignant masses merely contain solid components inside [28, 32]. Moreover, consistent growth of the cervical lesion during adulthood might indicate precancerosis and even cancerization [25]. Like the case reported herein, the growing size may also indicate the malignant nature or tendency of the lesion.

US is considered the first-line screening method for cervical cancer owing to its low price, convenience, and high reproducibility. Gray-scale US images of G-EAC show an enlarged cervix with multiple internal masses in the upper part of the cervix with multilocular cystic lesions, multilocular cystic lesions mixed with solid components, or purely solid lesions [10]. Occasionally, the lower part of the uterine body may be invaded. The multilobular cyst is commonly observed in gray-scale US, which shows multiple cystic or solid-cystic echoes of various sizes with smooth or blurred edges surrounded by larger cystic echoes. The color Doppler US can reveal the moderate or abundant blood flow signals of G-EAC [7,8,9,10, 33].

Interestingly, the three-dimensional reconstructed views of the G-EAC on 3D realisticVue markedly resemble the “cosmos pattern” in MRI [11]. Three-dimensional energy Doppler US can visualize the blood flow in the lesion and its spatial position in a three-dimensional structure. Moreover, it can measure the volume of the mass and display the vascularization index (VI), flow index (FI), and VFI [33], which are valuable for distinguishing between benign and malignant cervical masses, as malignant masses tend to have increased volume, uneven distribution of blood supply (some blood flows may increase and show a clump-like pattern), and increased VI, FI, and VFI values [34, 35]. CEUS can exhibit the underlying solid component inside of the cystic component, which is critical for distinguishing the benign or malignant nature of the cervical lesion, whereas the Gray-scale US cannot discriminate the malignant nature of the solid component within the cystic component. Under CEUS, the solid components of the mixed echo developed prior to the uterine myometrium, showing equal-high enhancement, whereas the normal cervical stroma does not develop before the myometrium, showing low-equal enhancement, which is one of the evident features for distinguishing the benign or malignant nature of lesions. Moreover, the CEUS provides valuable quantitative parameters and facilitates preoperative lesion assessment [36, 37]. In the current case, the quantitative parameters of CEUS, such as arrival time (AT), time to peak (TTP), rise time, and mean transit time (MTT), were shorter than those of the myometrium, and peak intensity (PI) was slightly lower than the myometrium, which was equally enhanced. Given that CEUS of the current case displayed a “quick-up and slow-down” pattern of solid components inside the mixed cervical echoes in the TIC, which noticeably differed from the “quick-up and quick-down” pattern of cervical carcinoma [36], we concluded that the cervical lesions in this case may be precancerous. These phenomena could be explained by the increased neovascularization within the lesion, gradual decrease or even absence of smooth muscle cells and elastic fibers, and the loss of function of vascular endothelial cells. Accordingly, the contrast agent fills and regresses faster than normal tissues, leading to shortened AT, TTP, rise time, MTT, and elevated PI [28, 38, 39]. In addition, the CEUS can present different aspects of local invasion, including parametrial extension and invasion of the vagina, uterine corpus, and other adjacent organs. CEUS has good concordance with MRI in evaluating the invasion of cervical cancer [40]. In this case, echoes of the parauterine and uterine body did not enhance synchronously with the solid components of the cervical mass, indicating that they were not invaded by the cervical lesion. Besides, the echoes of the serous layer were continuous, suggesting that the rectum and bladder were not invaded. MRI and traditional US have been reported to have low sensitivity in diagnosing local invasion of the cervical lesion into the vagina (merely 44.4% by MRI and 55.6% by traditional US [41]. Accordingly, another study reported that MRI and the traditional US were not entirely reliable in assessing vaginal invasion of the cervical lesion [42]. Therefore, we performed bi-plane transrectal US—an emerging technique that applies a high-frequency probe with a frequency range of 3.2–12.8 MHz and has a higher resolution compared to ordinary vaginal US probes. Thus, the entire structure of the vagina, including the anterior and posterior fornix and part of the cervix, could be depicted, and whether the vagina has been invaded could be clearly demonstrated.

However, US-obtained imaging features of G-EAC share many similarities with those of other cervical multicystic lesions, such as deep Nabothian cysts, tunnel cluster, cervicitis, and cervical endometrial hyperplasia, all of which exhibit an enlarged cervix with multicystic lesions, occasionally with increased blood flow signals in the presence of inflammation. Therefore, diagnosing G-EAC using solely gray-scale and color Doppler US remains a challenge. Nevertheless, it is possible to differentiate G-EAC from other cervical multicystic diseases mentioned above based on imaging features of the multimodal US and clinical manifestations with relatively high accuracy. Benign diseases such as Nabothian cysts, tunnel plexus, and cervical endometrial hyperplasia exhibit very few solid components in most cases, reduced or no blood flow signals in the cystic septum and wall, and a clear rim; their VI, FI, and VFI values are also low, typically small in size, with low risk of invading the lower part of the uterine body and they barely invade the parametrium and vagina. Therefore, a preliminary diagnosis of Nabothian cysts can be reached if the patient has the aforementioned medical imaging features and a history of dyspareunia or a feeling of fullness in the vagina. Tunnel clusters could be suspected if the patient has corresponding imaging characteristics and a history of multigravida, although the final diagnosis should be based on the results of histological examinations. Given the abundant inflammation-induced blood flow, uterine cervicitis may exhibit malignant signs in some cases with no “cosmos pattern” on MRI. Moreover, clinical manifestations may include pelvic pain and pressure, accompanied by a yellowish jelly-like vaginal discharge with an unpleasant odor. Following drug treatment, the blood flow signals may drop to normal levels during follow-up. Accordingly, cervicitis could be suspected based on these medical imaging features and clinical manifestations. In addition, a high T2 signal intensity and iso-intensity in T1 found in the cervix of patients of reproductive age or a history of oral progesterone usage may suggest cervical endometrial hyperplasia [30, 43]. However, special types of cervical hyperplasia, such as LEGH, are precancerous lesions of G-EAC. The preoperative differential diagnosis of LEGH is difficult and requires further diagnosis by biopsy or histopathological analysis. If the mass grows during follow-up, surgical intervention should be considered.

In the current case of PJS-correlated G-EAC, owing to these qualitative and quantitative features of multimodal US, G-EAC was suspected before surgery, and clinical staging of the lesions was performed to provide valuable clues for operative decision-making. By preoperatively suspecting G-EAC using multimodal US findings in this case of PJS-related G-EAC, we aimed to highlight the usefulness of multimodal US, which combines two- and three-dimensional US and CEUS, in the diagnosis of PJS-related G-EAC and create awareness to prevent missed diagnosis or misdiagnosis of G-EAC by sonologists. To the best of our knowledge, this is the first report describing the multimodal ultrasonic manifestations and the CEUS features of PJS-associated G-EAC, and additional cases and clinical evidence are required to verify the specificity of our findings.