What is the aim of the review?
We wanted to find out which imaging tests are better for identifying spread of disease among people with a first diagnosis of melanoma (primary staging) and among people with possible recurrence of melanoma (re-staging). We looked at the evidence for ultrasound, CT, MRI, and PET-CT and included 39 studies to answer these questions.
Why are imaging tests for melanoma important?
Melanoma is one of the most aggressive forms of skin cancer, with potential for metastases (cancer cells) to spread to the lymph nodes and other organs of the body. To make sure that people with melanoma receive the most appropriate and effective treatment, it is important to identify whether the disease has spread and to which parts of the body it has spread. This is called 'staging of disease'. Staging is done to find out if a melanoma has spread to regional lymph nodes or to lymph nodes close to the original melanoma, and to determine if the melanoma has spread to lymph nodes in other parts of the body or to organs of the body such as the liver or the brain (distant metastases). Imaging tests are tools that can be used to help find out how much the disease has spread. Several new treatments are now available for reducing the risk of spread of melanoma and for treating melanoma when it has spread.
What was studied in the review?
The review includes four imaging tests that create images of the body in different ways. Ultrasound uses high-frequency sound waves to create images, CT scans use ionising radiation in the form of X-rays (a very low dose of radiation), and MRI uses large magnets and non-ionising radiation in the form of radio waves (which are not harmful) to generate images of the body. PET-CT requires injection of a weakly radioactive substance (FDG). The PET part of the scan identifies areas of the body that take up a lot of FDG (indicating possibly cancerous cells), and the CT part of the scan helps to improve image quality and to more accurately pinpoint areas using more FDG. Ultrasound can also be performed along with a fairly simple procedure called 'fine needle aspiration cytology' (FNAC), by which a very fine needle is used to take a small sample of cells from a lymph node that looks suspicious on ultrasound. A microscope is then used to identify whether or not the cells are malignant.
Imaging can be used at different time points after diagnosis of melanoma. Healthcare providers can use imaging to look at the regional lymph nodes closest to the melanoma before a type of surgery called sentinel lymph node biopsy is performed. Sentinel lymph node biopsy takes out the lymph nodes that are most likely to have metastases inside them so they can be tested in a laboratory. Imaging can also be used after sentinel lymph node biopsy or in people with higher-risk melanoma to look for any spread of disease. Imaging can be used in people who were treated for melanoma at an earlier point and who might be having a recurrence of their disease.
What are the main results of the review?
Ultrasound of regional lymph nodes before sentinel lymph node biopsy
We found 11 relevant studies including 2614 people. Three of these studies compared ultrasound on its own to ultrasound combined with FNAC. Results suggest that the combined procedure correctly identifies around one-fifth of people with metastases in the lymph nodes with very few false positive results (people with incorrect diagnosis of metastasis). These results can be illustrated by imagining a group of 1000 people with melanoma who are going to have sentinel lymph node biopsy, of whom 237 (24%) have metastases in the lymph nodes. The combination of ultrasound with FNAC potentially allows 43 people with lymph node metastases to be identified and avoid a sentinel lymph node biopsy, at a cost of two people with false positive results who might go on to have the wrong treatment. Those with metastases in the lymph nodes that are missed on ultrasound (false negatives) will be identified on subsequent SLNB.
Whole body imaging (detection of any metastases)
We found 24 studies, but only nine were clear about the point in the time course of disease that imaging was carried out. Six studies including 492 people looked at imaging for primary staging following a confirmed diagnosis of melanoma, and three studies in 589 people evaluated re-staging of disease in people with possible recurrence of disease.
Most of the studies (6/9) considered PET-CT, two in comparison to CT alone, and three studies examined the use of ultrasound. We did not find any suitable studies of MRI in these groups.
Overall results suggest that PET-CT is better for correctly identifying people with metastatic spread of disease who might be having a recurrence of disease (re-staging) than people who have a new diagnosis of melanoma (primary staging). PET-CT also seems to be better than CT for identifying spread of disease in both groups of people, but studies were very small and results might not be reliable.
How reliable are the results of the studies included in this review?
In most of our studies, a reliable diagnosis of spread of disease (or reference standard) was made by performing biopsy and by following up with people over time using clinical assessment and imaging. There was often a lack of detail on how patients were followed up and which tests were used. Lots of studies did not include people at clearly defined time points in the disease process, making it difficult to assess the relevance of their results. Reporting of application and interpretation of tests was poor.
To whom do the results of this review apply?
Thirty-three studies were done in Europe (85%), and the rest in North America (n = 4), Asia (n = 1), or Oceania (n = 1). The average age of people in the studies was between 50 and 67 years, and around half were men. Studies mostly included people with melanoma on any part of the body, but two included only people with melanoma on the head or neck. Studies often included people at different stages of disease, and we were not able to look at the accuracy of tests for people at any particular disease stage. Studies were small, and their results might not match what happens in real life.
What are the implications of this review?
Reviewers found some evidence to support the use of imaging with ultrasound combined with FNAC before sentinel lymph node biopsy, but further work is needed to establish cost-effectiveness. Limited evidence is available for whole body imaging for primary staging or re-staging of disease. Available evidence is focused on PET-CT; there are few comparisons with CT and no comparisons with MRI. Future research needs to look at more than test accuracy and must consider the effects of different imaging tests on treatment decisions for patients.
How up-to-date is this review?
The reviewers searched for and included studies published up to August 2016.*
*In these studies, biopsy and clinical or imaging follow-up were the reference standards (methods of establishing the final diagnosis).
Review authors found a disappointing lack of evidence on the accuracy of imaging in people with a diagnosis of melanoma at different points on the clinical pathway. Studies were small and often reported data according to the number of lesions rather than the number of study participants. Imaging with ultrasound combined with FNAC before SLNB may identify around one-fifth of those with nodal disease, but confidence intervals are wide and further work is needed to establish cost-effectiveness. Much of the evidence for whole body imaging for primary staging or re-staging of disease is focused on PET-CT, and comparative data with CT or MRI are lacking. Future studies should go beyond diagnostic accuracy and consider the effects of different imaging tests on disease management. The increasing availability of adjuvant therapies for people with melanoma at high risk of disease spread at presentation will have a considerable impact on imaging services, yet evidence for the relative diagnostic accuracy of available tests is limited.
Melanoma is one of the most aggressive forms of skin cancer, with the potential to metastasise to other parts of the body via the lymphatic system and the bloodstream. Melanoma accounts for a small percentage of skin cancer cases but is responsible for the majority of skin cancer deaths. Various imaging tests can be used with the aim of detecting metastatic spread of disease following a primary diagnosis of melanoma (primary staging) or on clinical suspicion of disease recurrence (re-staging). Accurate staging is crucial to ensuring that patients are directed to the most appropriate and effective treatment at different points on the clinical pathway. Establishing the comparative accuracy of ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET)-CT imaging for detection of nodal or distant metastases, or both, is critical to understanding if, how, and where on the pathway these tests might be used.
Primary objectives
We estimated accuracy separately according to the point in the clinical pathway at which imaging tests were used. Our objectives were:
• to determine the diagnostic accuracy of ultrasound or PET-CT for detection of nodal metastases before sentinel lymph node biopsy in adults with confirmed cutaneous invasive melanoma; and
• to determine the diagnostic accuracy of ultrasound, CT, MRI, or PET-CT for whole body imaging in adults with cutaneous invasive melanoma:
○ for detection of any metastasis in adults with a primary diagnosis of melanoma (i.e. primary staging at presentation); and
○ for detection of any metastasis in adults undergoing staging of recurrence of melanoma (i.e. re-staging prompted by findings on routine follow-up).
We undertook separate analyses according to whether accuracy data were reported per patient or per lesion.
Secondary objectives
We sought to determine the diagnostic accuracy of ultrasound, CT, MRI, or PET-CT for whole body imaging (detection of any metastasis) in mixed or not clearly described populations of adults with cutaneous invasive melanoma.
For study participants undergoing primary staging or re-staging (for possible recurrence), and for mixed or unclear populations, our objectives were:
• to determine the diagnostic accuracy of ultrasound, CT, MRI, or PET-CT for detection of nodal metastases;
• to determine the diagnostic accuracy of ultrasound, CT, MRI, or PET-CT for detection of distant metastases; and
• to determine the diagnostic accuracy of ultrasound, CT, MRI, or PET-CT for detection of distant metastases according to metastatic site.
We undertook a comprehensive search of the following databases from inception up to August 2016: Cochrane Central Register of Controlled Trials; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists as well as published systematic review articles.
We included studies of any design that evaluated ultrasound (with or without the use of fine needle aspiration cytology (FNAC)), CT, MRI, or PET-CT for staging of cutaneous melanoma in adults, compared with a reference standard of histological confirmation or imaging with clinical follow-up of at least three months' duration. We excluded studies reporting multiple applications of the same test in more than 10% of study participants.
Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2)). We estimated accuracy using the bivariate hierarchical method to produce summary sensitivities and specificities with 95% confidence and prediction regions. We undertook analysis of studies allowing direct and indirect comparison between tests. We examined heterogeneity between studies by visually inspecting the forest plots of sensitivity and specificity and summary receiver operating characteristic (ROC) plots. Numbers of identified studies were insufficient to allow formal investigation of potential sources of heterogeneity.
We included a total of 39 publications reporting on 5204 study participants; 34 studies reporting data per patient included 4980 study participants with 1265 cases of metastatic disease, and seven studies reporting data per lesion included 417 study participants with 1846 potentially metastatic lesions, 1061 of which were confirmed metastases. The risk of bias was low or unclear for all domains apart from participant flow. Concerns regarding applicability of the evidence were high or unclear for almost all domains. Participant selection from mixed or not clearly defined populations and poorly described application and interpretation of index tests were particularly problematic.
The accuracy of imaging for detection of regional nodal metastases before sentinel lymph node biopsy (SLNB) was evaluated in 18 studies. In 11 studies (2614 participants; 542 cases), the summary sensitivity of ultrasound alone was 35.4% (95% confidence interval (CI) 17.0% to 59.4%) and specificity was 93.9% (95% CI 86.1% to 97.5%). Combining pre-SLNB ultrasound with FNAC revealed summary sensitivity of 18.0% (95% CI 3.58% to 56.5%) and specificity of 99.8% (95% CI 99.1% to 99.9%) (1164 participants; 259 cases). Four studies demonstrated lower sensitivity (10.2%, 95% CI 4.31% to 22.3%) and specificity (96.5%,95% CI 87.1% to 99.1%) for PET-CT before SLNB (170 participants, 49 cases). When these data are translated to a hypothetical cohort of 1000 people eligible for SLNB, 237 of whom have nodal metastases (median prevalence), the combination of ultrasound with FNAC potentially allows 43 people with nodal metastases to be triaged directly to adjuvant therapy rather than having SLNB first, at a cost of two people with false positive results (who are incorrectly managed). Those with a false negative ultrasound will be identified on subsequent SLNB.
Limited test accuracy data were available for whole body imaging via PET-CT for primary staging or re-staging for disease recurrence, and none evaluated MRI. Twenty-four studies evaluated whole body imaging. Six of these studies explored primary staging following a confirmed diagnosis of melanoma (492 participants), three evaluated re-staging of disease following some clinical indication of recurrence (589 participants), and 15 included mixed or not clearly described population groups comprising participants at a number of different points on the clinical pathway and at varying stages of disease (1265 participants). Results for whole body imaging could not be translated to a hypothetical cohort of people due to paucity of data.
Most of the studies (6/9) of primary disease or re-staging of disease considered PET-CT, two in comparison to CT alone, and three studies examined the use of ultrasound. No eligible evaluations of MRI in these groups were identified. All studies used histological reference standards combined with follow-up, and two included FNAC for some participants. Observed accuracy for detection of any metastases for PET-CT was higher for re-staging of disease (summary sensitivity from two studies: 92.6%, 95% CI 85.3% to 96.4%; specificity: 89.7%, 95% CI 78.8% to 95.3%; 153 participants; 95 cases) compared to primary staging (sensitivities from individual studies ranged from 30% to 47% and specificities from 73% to 88%), and was more sensitive than CT alone in both population groups, but participant numbers were very small.
No conclusions can be drawn regarding routine imaging of the brain via MRI or CT.