Magnetic Resonance Imaging findings in Primary Extracutaneous Melanoma from a Tertiary Cancer Care Centre: A Retrospective Cohort Study
Correspondence Address :
Dr. Renuka Gopalakrishnan,
Manjusha, Puthupally Lane, Medical College PO, Thiruvananthauram-695011, Kerala, India.
E-mail: drrenukag@gmail.com
Introduction: Extracutaneous melanomas are rare, aggressive type of tumour, clinically and biologically distinct from their cutaneous counterpart. The two large broad categories of the extracutaneous melanomas are ocular and mucosal subtypes. Melanomas are classically hyperintense on T1 weighted images and hypointense on T2 weighted images due to the paramagnetic effects of melanin and presence of paramagnetic elements.
Aim: To describe the Magnetic Resonance Imaging (MRI) findings in primary extracutaneous melanomas at various anatomic sites.
Materials and Methods: This was a retrospective cohort study in which 13 cases of primary extracutaneous melanomas were identified from Picture Archiving and Communication System (PACS) archive over a period of eight years (January 2013-December 2020). Location and morphology of the tumour, signal intensity characteristics in T1 weighted, T2 weighted (hyperintense/isointense/hypointense to adjacent muscle) and Diffusion Weighted Imaging (DWI) (presence or absence of diffusion restriction) were analysed.
Results: The ocular melanomas (n=2) were seen as well-defined small intraocular mass attached to the choroid. The mucosal melanomas of the nasal cavity (n=2), rectum (n=4), vagina (n=3) and cervix (n=2) presented as large intraluminal polypoidal masses. Three categories of MRI appearances emerged in this study. Majority of the cases (n=8) showed hyperintense signals in T1-weighted images and hypointense signals in T2 weighted images (category 1). Diffusion restriction was seen in all cases (n=13) with low Apparent Diffusion Coefficient (ADC) values which ranged from 439-966 mm/sec with an average value of 755 mm/sec.
Conclusion: T1 and T2 shortening are typical of melanoma, the absence of these does not exclude the diagnosis. Although diffusion restriction and low ADC values help in the diagnosis of these tumours, they do not tend to play a specific role in the diagnosis.
Apparent diffusion coefficient, Diffusion weighted imaging, T1 and T2 weighted imaging
Malignant melanoma develops from melanocytes derived from neural crest cells. They usually occur in the skin (cutaneous melanoma), but they can also occur in any organ with melanin containing cells. Only 4-5% of primary melanomas are extracutaneous of which 70% are ocular and remaining are the mucosal subtype (1),(2). Extracutaneous melanomas are rare, aggressive type of tumour clinically and biologically distinct from the cutaneous counterpart. The two large broad categories of the extracutaneous melanomas are ocular and mucosal subtypes. Ocular melanoma is the most common primary eye tumour in adults and second most common type of melanoma after cutaneous subtype (2). Mucosal melanoma arises from different mucosal lined organs; which include sinonasal tract, oral cavity, gastrointestinal tract (oesophagus, anorectum and biliary) and genito-urinary tract (vulva, vagina, cervix and urethra). Mucosal melanomas tend to occur in elderly patients with a median age at diagnosis of 70 years, having no clear predisposing factors and more prevalent in females; possibly due to increased incidence of vulvovaginal melanoma. They are diagnosed late and associated with poor outcome with an overall survival rate of 25% compared to 81% with cutaneous melanoma (3),(4).
Extracutaneous melanomas tend to metastasise to various organs like lymph node, bone, lung, liver, spleen, gastrointestinal tract and subcutaneous tissue (4). Melanomas are classically hyperintense on T1 weighted images and hypointense on T2 weighted images due to the paramagnetic effects of melanin and presence of paramagnetic elements, including copper, manganese and zinc (5),(6),(7). Diffusion restriction with low ADC values have also been described. The purpose of this study was to describe the imaging findings in primary extracutaneous melanomas at various anatomic sites and look for any specific finding in MRI which would be helpful for its diagnosis by retrospectively evaluating the imaging archives of institution.
This retrospective cohort study was conducted in the Department of Radiodiagnosis, affiliated to a tertiary cancer centre in Southern India (Regional Cancer Centre, Thiruvananthapuram, Kerala, India). MRI scans done between January 2013-December 2020 were retrieved from PACS. Institutional Review Board (IRB) approval was obtained for this study (IRB no 07/2021/04). Written consent was waived. Data collection and analysis was done over a period of one month after obtaining IRB clearance. A total of 1400 melanomas (cutaneous and extracutaneous) were reported from this tertiary cancer care centre during the study period, mostly of cutaneous origin. Authors retrospectively identified cases from PACS system (GE centricity) to select subjects for this study. Out of the 78 cases of primary extracutaneous melanomas, 13 cases diagnosed with primary extracutaneous melanoma in this centre were identified for this study.
Inclusion criteria: Patients with primary extracutaneous melanoma who had initial MRI workup done at this centre with histological proven cases were included in the study.
Exclusion criteria: The patients who had their initial imaging done elsewhere or who had technically poor quality scans were excluded (n=65).
Imaging
The MRI was performed on 1.5 Tesla machine (GE Health care Milwaukee WI). In MRI the following features were analysed: tumour location, shape, signal intensity characteristics, and ADC value on DWI sequences. In addition other features like infiltration of surrounding tissues and lymphadenopathy were also noted. Each MRI examination was evaluated by two radiologists to determine the signal intensity of the lesion on T1-weighted and T2-weighted imaging when compared with adjacent musculature and categorised as hypointense/isointense/hyperintense signals relative to muscle in each sequence. Cases which showed hyperintense signals on T1-weighted images and hypointense signals on T2 weighted images (category 1). Cases which showed hypointense signals on T1-weighted images and hyperintense signal on T2 weighted images (category 2). Cases which showed hyperintense signals on both T1 and T2 weighted images (category 3).
Statistical Analysis
Descriptive statistics with analysis of frequency and percentage of various parameters was done.
The age of presentation in this study cohort ranged from 33-80 years with a mean age of 62±10 years. This study population showed a slight female preponderance, seven females and six males, possibly due to inclusion of female genital tract melanomas. Primary location of the tumour were in the choroid (n=2), nasal cavity (n=2) rectum (n=4), vagina (n=3) and cervix (n=2).
Tumour location and morphology: The ocular lesions were seen as well-defined small intraocular mass attached to the choroid. No extraocular extension was seen (Table/Fig 1)a-d. Nasal mucosal melanomas were seen as polypoid mass arising from the lateral nasal wall (Table/Fig 2)a,b. Bone destruction and lymph node enlargement were seen in one patient. Rectal melanomas were seen as large polypoid non obstructing intraluminal masses (Table/Fig 3)a-c. Perirectal infiltration was seen in one patient and lymphadenopathy in three patients (Table/Fig 3)a,b, (Table/Fig 4)a-d. Out of the three patients with vaginal melanoma, only one showed infiltration into the surrounding fat, remaining two were confined to vaginal wall with intraluminal component (Table/Fig 5) a-d. Melanoma of cervix were also seen as large intraluminal polypoid mass (Table/Fig 6)a,b and (Table/Fig 7)a,b. Neither parametrial infiltration nor lymphadenopathy was seen any of the cases with melanoma of cervix.
Imaging findings: Three categories of MRI appearances were demonstrated in this study. Majority of the cases (n=8) showed hyperintense signals on T1-weighted images and hypointense signals on T2 weighted images (category 1). Four cases showed hypointense signals on T1-weighted images and hyperintense signal on T2 weighted images (category 2).The remaining one case showed hyperintense signals on both T1 and T2 weighted images (category 3). Diffusion restriction was seen in all cases with low ADC values which ranged from 439-966 mm/sec with an average value of 755±201 mm/sec. The MRI signal characteristics of each lesion in various anatomical location is given in (Table/Fig 8).
Extracutaneous melanomas are rare tumours with poor prognosis. The incidence of extracutaneous melanoma in this study in comparison to other studies is mentioned in (Table/Fig 9) (1),(8),(9),(10),(11). Accurate diagnosis of this disease is required to prevent early metastatic spread (2). Treatment modalities usually are surgery, radiotherapy, chemotherapy and immunotherapy depending on the primary site of tumour and metastatic disease. According to various literature reviewed ocular subtype is the most common primary extracutaneous melanoma (2),(5). In this study, only two cases of ocular melanoma was observed. The other patients treated in this institution were those who were diagnosed and investigated at other ophthalmology centres and referred to us for treatment; hence were not included in this study. The median age at the time of diagnosis of non cutaneous melanoma was 56 years in this study which was similar to recent study by Lian B et al., which included 706 patients. In this study, there is slight female preponderance (54%) compared to male (46%) which was similar to that in literature, primarily because melanoma arising within the female genital tract is more common (2),(3),(4).
Traditionally melanomas tend to shorten T1 and T2 relaxation time due to its paramagnetic effect, resulting in hyperintense signals on T1 weighted imaging and hypointense signals on T2 weighted imaging (5),(6),(7). Signal intensities can be correlated with percentage of melanin containing cells in tumour; greater the concentration of melanin, greater the high signal intensity on T1 weighted images (7),(12). In the current study majority of the cases (62%) showed typical appearance of melanoma. A total of 31% of the cases showed signal intensity shown by other tumours, i.e. hypointense signals on T1 weighted imaging and hyperintense signals on T2 weighted imaging and remaining appeared hyperintense in both T1 and T2 weighted sequences. The study by Marx HF et al., describing the MRI features in melanoma also showed similar signal intensity patterns (6). Their study proposed various explanations for the other spectrum of imaging findings other than the classical one such as differing concentrations of melanin, propensity for hemorrhage, use of wide range of Repetition Time (TR) and Time to Echo (TE) values and signal intensity of surrounding tissue (as in the case of choroid melanoma where the background vitreous humour will help to enhance the signal of the tumour).
Morphology of the tumour and MR signal characteristics of melanomas at various anatomical sites in this study are as follows:
Ocular: According to literature (5),(12), ocular melanoma most commonly occurs in uvea of which choroid melanoma is most common subtype. In the two cases of ocular melanoma in this study, mass was seen arising from the choroid lining. Both lesions appeared hyperintense on T1-weighted images hypointense on T2-weighted images. Case studies by Houle V et al., and Jiblawi A et al., on choroidal melanoma also showed similar MRI signal pattern (13),(14).
b3bNasal cavity: In this study, frequent site of origin of nasal cavity melanoma was lateral nasal wall (83%). Literature review showed that majority of the lesions rise in the lateral nasal wall and nasal septum followed by maxillary sinus (3),(15). Both the lesions in this study appeared hyperintense on T1-weighted images and hypointense on T2-weighted images. In a study of MRI findings in sinonasal melanoma on 12 patients by Yousem DM et al., 60% patients showed T1 hyperintensity (16). Signals on T2WI was variable in the study appearing hypo, iso or hypointense.
Rectum: All patients with rectal melanoma had presented with intraluminal large polypoid mass with focal luminal expansion and without bowel obstruction. This observation was similar to study done by Park HJ et al., evaluating the MRI findings of anorectal melanoma in 12 patients in which similar morphology was seen in all cases (17). In the present study, 25% cases of rectal melanomas showed hyperintense signals on T1-weighted images and hypointense on T2-weighted images as expected; remaining 75% cases showed hypointense signals on T1-weighted images and hyperintense on T2-weighted images. In the study by Park HJ et al., (17) T1 hyperintensity was seen in about 66% and 33% had iso intense signals, contrary to this study. Their study showed perirectal infiltration in all patients (100%) and lymph node enlargement in 75%; Perirectal infiltration and lymphadenopathy was seen in 75% of patients in current study. Diffusion restriction was seen in all their cases, similar to this study.
Vagina: Vaginal melanomas appeared as intraluminal polypoidal masses in this study. One of the patients showed infiltration of parametrial fat while other two remained confined to vaginal wall. Lymph node enlargement was not seen in any. In this study lesions appeared hyperintense in T1 weighted images and hypointense in T2 weighted images in 66%. In the study done by Takehara M et al., (18) in six patients with melanoma of female genital tract, 75% patients with vaginal melanoma had high signal intensity on T1-weighted images similar to this study. In case studies by Moon WK et al. and Kim H et al., (20) it was concluded that if the vaginal mass showed intermediate to high signal intensity on T1 weighted images, melanoma should be considered (19),(20).
Cervix: Melanoma of cervix in this study were seen as large polypoid lesions in the endocervical canal without invasion of stroma. Parametrial infiltration or lymphadenopathy was not seen. Both lesions showed hyperintense signals on T1-weighted images. But on T2 weighted images one lesion appeared hypointense as expected while the other appeared hyper intense. Only very few case reports have been published regarding melanoma of cervix. A case report by Lee JH et al., showed hyperintense mass in T1 weighted images with parametrial infiltration and bone metastasis (21). The study by Takehara M et al., also included a case with mass in cervix which had a high signal intensity on both T1 and T2-weighted images (18).
ADC: The ADC values were evaluated for all the 13 patients who underwent MRI evaluation. The values are as mentioned in (Table/Fig 8). The ADC values ranged from 439-966 mm/sec with an average value of 755 mm/sec. In the study by Erb-Eigner K et al., on 44 patient with ocular melanoma, it was found that melanomas tend to show low ADC value below 1000 mm/sec with a mean ADC value of 891 mm/sec (22). In the study by Liu QY et al., on patients with vaginal melanoma an average ADC value of 647 mm/sec was obtained (23). The ADC values obtained in this study was comparable to the values obtained in the above two studies.
Limitation(s)
Due to rareness of this disease entity only a small sample size could be included. As MRI findings were only analysed, this study had to exclude patients with buccal mucosal melanoma for which imaging modality used is Computed Tomography (CT) scan.
Melanomas become the most important differential in a large polypoidal intraluminal non obstructing mass with high signal intensity on T1 weighted images and low signal intensity on T2 weighted images in MR imaging in a mucosa lined organ. Signal patterns showing T1 and T2 shortening are typical of melanoma, the absence of these does not exclude the diagnosis. Although diffusion restriction and low ADC values help in the diagnosis of these tumors, they do not tend to play a specific role in the diagnosis. Because of rarity of this entity a multi-institutional study will be more useful for better understanding.
Authors would like to thank Dr. Jasmine Kaur Ahuja for her help.
DOI: 10.7860/JCDR/2022/55271.16347
Date of Submission: Jan 30, 2022
Date of Peer Review: Mar 16, 2022
Date of Acceptance: Apr 02, 2022
Date of Publishing: May 01, 2022
AUTHOR DECLARATION:
• Financial or Other Competing Interests: None
• Was Ethics Committee Approval obtained for this study? Yes
• Was informed consent obtained from the subjects involved in the study? NA
• For any images presented appropriate consent has been obtained from the subjects. NA
PLAGIARISM CHECKING METHODS:
• Plagiarism X-checker: Feb 02, 2022
• Manual Googling: Mar 04, 2022
• iThenticate Software: Apr 01, 2022 (12%)
ETYMOLOGY: Author Origin
- Emerging Sources Citation Index (Web of Science, thomsonreuters)
- Index Copernicus ICV 2017: 134.54
- Academic Search Complete Database
- Directory of Open Access Journals (DOAJ)
- Embase
- EBSCOhost
- Google Scholar
- HINARI Access to Research in Health Programme
- Indian Science Abstracts (ISA)
- Journal seek Database
- Popline (reproductive health literature)
- www.omnimedicalsearch.com