1. Nuclear Medicine and PET Department, St Vincent’s Hospital, Sydney, New South Wales.
2. University of New South Wales, Sydney, New South Wales.
This review discusses the relevant imaging techniques for both melanoma and non-melanoma skin cancers, including basal cell carcinomas, cutaneous squamous cell carcinomas and Merkel cell carcinomas. The diagnostic value of sentinel lymph node mapping, CXR, CT, PET, PET/CT and MRI are discussed, and their role for each stage and type of cutaneous malignancies considered. There are currently no recommendations for the use of diagnostic CT or PET /CT for initial staging of low risk melanoma patients around the time of diagnosis. Both PET and PET/CT have been shown to change the management of many patients with advanced melanoma, and it is recommended that FDG-PET be performed prior to resection of regional or distant metastatic deposits. MRI plays an important adjunctive role in the assessment of brain metastases in melanoma, and for assessment of perineural invasion in non melanoma skin cancers. Cumulative doses of radiation to patients for staging and surveillance imaging, and the life expectancy of the individual patient, should be factored into any decisions regarding what scans are appropriate for them.
Recommendations for appropriate imaging in patients with melanoma rely heavily on the patient’s disease status. In patients with stage II melanoma, there is a documented role for sentinel lymph node mapping using lymphoscintigraphy,1 but no role for further imaging procedures in the absence of symptoms. However, the case is different in patients with higher risk tumours and those with recurrent disease, who may benefit from a range of imaging procedures.
Lymphatic mapping and sentinel lymph node biopsy are recommended for disease staging in patients with high risk stage I/II disease.1-3 Lymphoscintigraphy involves the injection of a small volume of radiolabelled colloid around the tumour site, and then imaging to determine which regional node field contains the sentinel node or nodes (a sentinel node being defined as any node receiving direct drainage from a tumour site).4,5 Imaging is important, as often more than one regional node field is involved 4, and for accurate staging all sentinel lymph nodes should be pathologically examined. Lymphoscintigraphy identifies 94.5% of sentinel nodes.6 The use of SPECT (Tomographic nuclear imaging) / CT, in addition to standard dynamic lymphoscintigraphy, further improves diagnostic accuracy.7-9 Two studies have undertaken direct comparisons between lymphoscintigraphy and PET imaging, as a staging procedure in early stage melanoma. These found that lymphoscintigraphy was vastly diagnostically superior to PET imaging, with a sensitivity of 14% for PET imaging in both studies and a sensitivity of 86-100% for lymphoscintigrapy.10,11
In early stage melanoma, sensitivities of just 17% have been found in a number of meta-analyses and pooled study results assessing the diagnostic value of PET and PET/CT in early stage melanoma.6, 12 This is not surprising, as sentinel lymph node involvement with metastatic melanoma is most commonly microscopic.13 The sensitivity of PET/CT is limited by the volume of disease present, with the sensitivity for metastatic and lymph node disease dropping off significantly with tumour deposits of < 78mm3 in size (4mm diameter). PET has a sensitivity of < 50% for lesions < 4mm in size (80mm3 ).14 Hence, there are currently no recommendations for the use of PET/CT or diagnostic CT for initial staging of low risk melanoma patients around the time of diagnosis.15
PET and PET/CT
Metastatic melanoma deposits characteristically have high metabolic activity, showing up distinctly in 18F-FDG-PET/CT imaging. While FDG-PET/CT has not been shown to be diagnostically useful in early stage (stage I and II) melanoma where lymph node disease, if present, is usually of low volume and below the resolution limits for PET or CT technology, it has an important diagnostic role in patients with stage III and IV melanoma. This change in sensitivity, based on the stage of disease, was demonstrated elegantly by Wagner et al, who found a PET sensitivity of 0% in stage I disease, 24% in stage II disease, 81% in stage III disease and 100% in stage IV disease in patients with cutaneous melanoma.16 Multi study reviews of PET have found similarly high diagnostic accuracy values for identifying metastatic melanoma in stage III and IV disease. A pooled analysis of 753 patients with stage III and IV melanoma showed that F18– FDG-PET had a pooled sensitivity of 88%, a specificity of 82% and an accuracy of 86%.6
The diagnostic accuracy of PET/CT is significantly higher, with fewer false positive and false negatives than FDG-PET alone.17 It is now routine to undertake combined PET/CT imaging that is simultaneously or concurrently acquired on the same camera. The CT component can be either multislice contrast-enhanced, or a low radiation dose CT. A study of 50 patients with metastatic melanoma compared the diagnostic accuracy of contrast-enhanced diagnostic CT, PET and PET/CT (both contrast-enhanced and non-contrast low dose). The authors reported a sensitivity and specificity for diagnostic contrast-enhanced CT of 85% and 63 %, for PET of 90% and 88%, for PET/non diagnostic CT of 97% and 93% and for PET/diagnostic-contrast-enhanced CT of 100% and 93% respectively.18 It is interesting to note that while the addition of CT information to PET significantly improved sensitivity and specificity, there was little additional improvement in sensitivity and no improvement in specificity when comparing low-dose CT and contrast-enhanced CT.
Both PET and PET/CT have been shown to change the management of many patients with advanced melanoma, and it is recommended that FDG-PET be performed prior to resection of regional or distant metastatic deposits.6 Etchebehere et al assessed the ability of PET/CT to change management in 78 patients with locoregional or distant recurrence of melanoma.19 PET/CT changed management in 27% of the group studied.
There are a number of limitations to FDG-PET imaging that should be taken into account in evaluating scan results. High background activity in normal brain tissue reduces the sensitivity of PET for the detection of melanoma brain metastases. This mandates that either diagnostic CT of the brain or MRI be undertaken in patients at high risk of melanoma brain metastases. For technical reasons, the intensity of FDG uptake in small lung metastases may often be reduced, particularly at the lung bases. The combined technique PET/CT is particularly useful in the detection of small lung metastases, often missed on PET alone.12 As FDG-PET measures glucose uptake, a low false positive rate due to scan findings related to inflammation, sarcoidosis and unrelated tumours, is inevitable.
Although it is a widely used technique for staging and surveillance, there have been relatively few studies assessing the diagnostic value of contrast-enhanced CT in patients with melanoma 6. The technique suffers the same limitations as FDG–PET in early stage melanoma as it requires anatomical distortion for detection. A meta-analysis of diagnostic imaging modalities in later stage melanoma found a significantly higher diagnostic accuracy for PET/CT in disease surveillance (86% sensitivity, 91% specificity) than for CT alone (sensitivity 63%, specificity 71%).20 In fact, no study has found the diagnostic accuracy of contrast-enhanced CT to be higher than PET or PET/CT for cutaneous malignancies.
Magnetic Resonance Imaging
MRI plays an important adjunctive imaging role in the management of patients with cutaneous melanoma, particularly those with suspected or documented brain metastases. MRI is significantly more sensitive than CT for the detection of metastatic disease in the brain,21, 22 and also provides more detailed information about possible involvement of the spinal cord and leptomeninges. Contrast enhanced CT scanning is widely used because of its ready accessibility and relatively low cost. However, brain MRI for patients with primary cancers that frequently metastasise to the brain (including melanoma) is probably cost effective. Numerous studies have shown that contrast-enhanced MRI detects two to three times as many lesions as contrast-enhanced CT, especially lesions less than 5mm in diameter. In addition, approximately 20% of patients with solitary metastatic lesions in the brain on CT show multiple lesions on MRI.21-25
Plain chest X-ray
Routine surveillance of patients at higher risk of melanoma recurrence with plain chest x-ray (CXR) has been recommended in treatment guidelines until recently. However, a review of 1235 patients with melanomas >1mm in Breslow thickness, followed for a median of 74 months, found that the sensitivity of surveillance CXR was 7.7% and the specificity 96.5%. Of those diagnosed with metastatic disease on CXR (0.9%), only 0.2% had isolated pulmonary metastases amenable to resection. Hence, CXR is no longer recommended for standard surveillance in melanoma patients.26,27
Merkel cell carcinoma
Merkel cell carcinoma is a rare, aggressive skin malignancy with a high recurrence rate in regional node fields. Unlike in melanoma, the role of lymphoscintigraphy and sentinel lymph node biopsy is controversial in the management of Merkel cell carcinoma. Lymphoscintigraphy sensitivity in Merkel cell carcinoma varies in the literature from just 27% to 32%.28, 29 However, a positive sentinel node has a high predictive value for relapse in the regional node field, and is an indication for adjuvant radiotherapy treatment 28, 30. Unfortunately, a negative sentinel node biopsy in Merkel cell cancer does not preclude early recurrence in the same lymph node field. The use of F18-FDG-PET in the staging of Merkel cell cancer is also yet to be established, with no large prospective studies to date. A retrospective study evaluating the use of diagnostic CT, PET/CT and MRI in the initial lymph node staging of 99 patients with Merkel cell carcinoma found a sensitivity of 85% and specificity of 95% with PET, 47% and 97% for CT, and 0% and 86% for MRI (histopathology of lymph nodes was used for determining sensitivity and specificity)31. A further small study of 18 patients found that PET had a significant management impact in patients with Merkel cell cancer, altering staging in 33% and changing management in 43%.32 While Merkel cell cancer is a form of neuroendocrine tumour, it tends to have a high mitotic rate, but imaging with radiolabelled Somatostatin analogues has not been shown to be more sensitive than 18F FDG-PET imaging.33
Basal and squamous cell carcinoma
Basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) constitute approximately 95% of non-melanoma skin cancers.34, 35 The majority of these cancers do not require diagnostic evaluation. However, high risk patients, particularly with recurrent or invasive tumours, tumours at risk for regional or distant metastasis or accompanied by clinical signs of perineural involvement, warrant imaging to assess morphology of the primary tumour site, as well as to identify distant metastatic disease.36
Given its superior soft tissue resolution, MRI is considered to be the imaging modality of choice for evaluation of the primary tumour site, with excellent depiction of the extent of locoregional disease. In certain locations it is particularly important in presurgical planning.37 Given the tendency of some non-melanoma skin cancers to spread by perineural invasion, MRI is the modality of choice.36, 38 Nemzek and colleagues reported a 95% sensitivity for MRI detection of perineural invasion.38 MRI is also superior in identifying intracranial metastases and intracranial extension, including meningeal involvement, as well as subtle marrow infiltration. Diffusion-weighted MRI has become a promising biomarker for assessing tumour response to therapy.39
For evaluation of nodal involvement and distant metastatic disease, molecular imaging with fusion PET-CT and diagnostic CT are the imaging modalities of choice.34The sensitivity and specificity of PET/CT in staging and follow up for aggressive non-melanoma skin cancers remain to be elucidated. Over the last decade, fusion PET/CT has become an established, powerful imaging modality in the field of oncology, providing functional and anatomical correlation. However, in non-melanoma skin cancers, few data are available regarding staging and follow-up evaluation. Boswell and colleagues reported cases with metastatic BCC to the lung that was detected on PET/CT.40 In tumour of unknown origin, combined PET/CT has gained wide acceptance.41 A recent meta-analysis with encouraging data showed that, overall, FDG-PET/CT is able to detect 37% of primary tumours in patients with cancers from an unknown primary, with both sensitivity and specificity of 84%.42 Gourin and colleagues reported detection of distant metastatic disease in 15 of 64 patients, 13 of which were unsuspected prior to PET/CT.43 In patients with head and neck cancers receiving radiotherapy, a negative PET/CT result within six months after radiotherapy correlated with statistically significantly improved two-year overall survival rates in a study by Kao and colleagues, who followed 80 patients over a median of 21 months.44
Given its ready availability and rapid acquisition times, diagnostic CT has been used widely for routine surveillance. One advantage of CT over MRI is the increased sensitivity in the detection of subtle cortical bony erosion,34 but for small lesions or pathological changes in normal-sized tissues can be missed by CT.41
Cumulative doses of radiation to patients for staging and surveillance imaging must be factored into any decisions regarding which scans are appropriate. Given that the diagnostic yield of either CT or PET/CT is low in melanoma patients with AJCC stage I or II disease, it is not recommended that diagnostic CT, PET/CT or CXR be used in the absence of symptoms requiring investigation. By contrast, lymphoscintigraphy has a high diagnostic accuracy and delivers a very low radiation dose to this group of patients with a better prognostic outcome. In patients with AJCC stage III or IV melanoma, life expectancy is considerably reduced, and the diagnostic value of regular imaging must be weighed up carefully against the relatively low lifetime risk of secondary cancers induced by frequent imaging procedures.
PET imaging now almost always incorporates a low dose CT for the purposes of attenuation correction and anatomical detailing. This low dose CT is weight dependent in terms of radiation delivered and gives an average dose of 4mSv per patient in our institution (range 2.9-9.6 mSv). The 18F- FDG delivers 5-7 mSv per patient, giving an average radiation dose of 9 mSv per patient for a PET/ low dose CT scan. A comparative study of radiation doses across institutions found similar results for low dose PET/CT images. They also found that, on average, a diagnostic CT scan added an extra 14-19 mSv to the procedure.45 Putting this in perspective, the expected background radiation dose for a person living in Australia is 2mSv/year.
Rapidly evolving technology in imaging sometimes makes choosing the most appropriate imaging procedure for an individual patient difficult. MRI and lymphoscintigraphy have proven valuable in local disease characterisation and regional lymph node involvement, while PET/CT is proving the most diagnostically accurate procedure for assessment of distant metastatic disease. The clinician must take into account both the stage and type of cutaneous malignancy in deciding which imaging technique to employ, or indeed whether imaging is required at all.
1. Wong SL , Hurley P, Lyman GH. Sentinel lymph node biopsy for melanoma: American Society of Clinical Oncology and Society of Surgical Oncology Joint Clinical Practice Guideline. J Oncol Pract. 2012; 8(4):e65-e66.
2. Gershenwald JE, Coit DG, Sondak VK, Thompson JF. The challenge of defining guidelines for sentinel lymph node biopsy in patients with thin primary cutaneous melanomas. Ann Surg Oncol. 2012;91(11):3301-3.
3. Murali R SR, Thompson RF. Can we better identify thin cutaneous melanomas that are likely to metastasize and cause death? Ann Surg Oncol. 2012;19)11):3310-2.
4. Uren RF, Howman-Giles R, Chung D, Thompson JF. Guidelines for lymphoscintigraphy and F18 FDG PET scans in melanoma. J Surg Oncol. 2011;104(4):405-19.
5. Morton DL, Wen DR, Wong JH, Economou JS, Cagle LA, Storm FK, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 1992; 127(4):392-9.
6. Ho Shon IA, Chung DK, Saw RP, Thompson JF. Imaging in cutaneous melanoma. Nucl Med Commun. 2008;29(10):847-76.
7. Even-Sapir E, Lerman H, Lievshitz G, Khafif A. Lymphoscintigraphy for sentinel node mapping using a hybrid SPECT/CT system. J Nucl Med. 2003;44(9):1413-20.
8. Kretschmer L, Altenvoerde G, Meller J, Zutt M, Funke M, Neumann C, et al. Dynamic lymphoscintigraphy and image fusion of SPECT and pelvic CT-scans allow mapping of aberrant pelvic sentinel lymph nodes in malignant melanoma. Eur J Cancer. 2003;39(2):175-83.
9. Uren RF. SPECT/CT Lymphoscintigraphy to locate the sentinel lymph node in patients with melanoma. Ann Surg Oncol. 2009;16(6):1459-60.
10. Belhocine T, Pierard G, De Labrassinne M, Lahaye T, Rigo P. Staging of regional nodes in AJCC stage I and II melanoma: 18FDG PET imaging versus sentinel node detection. Oncologist 2002;7(4):271-8.
11. Singh B, Ezziddin S, Palmedo H, Reinhardt M, Strunk H, Tuting T, et al. Preoperative 18F-FDG-PET/CT imaging and sentinel node biopsy in the detection of regional lymph node metastases in malignant melanoma. Melanoma Res. 2008;18(5):346-52.
12. Belhocine TZ, Scott AM, Even-Sapir E, Urbain JL, Essner R. Role of nuclear medicine in the management of cutaneous malignant melanoma. J Nucl Med. 2006;47(6):957-67.
13. Sanki A, Uren RF, Moncrieff M, Tran KL, Scolyer RA, Lin HY, Thompson JF. Targeted high-resolution ultrasound is not an effective substitute for sentinel lymph node biopsy in patients with primary cutaneous melanoma. J Clin Oncol. 2009;27(33):5614-9.
14. Mijnhout GS, Hoekstra OS, van Lingen A, van Diest PJ, Ader HJ. How morphometric analysis of metastatic load predicts the (un)usefulness of PET scanning: the case of lymph node staging in melanoma. J Clin Pathol. 2003;56(4):283-6.
15. Network AC [Internet]. Canberra: Clinical practice guidelines for the management of melanoma in Australia and New Zealand. [updated 20 February 2012; cited 23 July 2012]. Available from: http://www. nhmrc.gov.au/guidelines/publication/cp111.
16. Wagner JD, Schauwecker DS, Davidson D, Wenck S, Jung SH, Hutchins G. FDG-PET sensitivity for melanoma lymph node metastases is dependent on tumor volume. J Surg Oncol. 2001;77(4):237-42.
17. Reinhardt MJ, Joe AY, Jaeger U, Huber A, Matthies A, Bucerius J, et al. Diagnostic performance of whole body dual modality 18F-FDG PET/CT imaging for N- and M-staging of malignant melanoma: experience with 250 consecutive patients. J Clin Oncol. 2006;24(7):1178-87.
18. Pfluger T, Melzer HI, Schneider V, La Fougere C, Coppenrath E, Berking C, et al. PET/CT in malignant melanoma: contrast-enhanced CT versus plain low-dose CT. Eur J Nucl Med Mol Imaging. 2011;38(5):822-31.
19. Etchebehere EC, Romanato JS, Santos AO, Buzaid AC, Camargo EE. Impact of [F-18] FDG-PET/CT in the restaging and management of patients with malignant melanoma. Nucl Med Commun. 2010;31(11):925-30.
20. Xing Y, Bronstein Y, Ross MI, Askew RL, Lee JE, Gershenwald JE, et al. Contemporary diagnostic imaging modalities for the staging and surveillance of melanoma patients: a meta-analysis. J Natl Cancer Inst. 2011;103(2):129-42.
21. Yock DH. Magnetic Resonance Imaging of CNS Disease: A Teaching File. St Louis: Mosby-Year Book Inc; 1995.
22. Mohr P EM, Hauschild A. Staging of Cutaneous Malignancy. Ann Oncol. 2009;20(6):14-21.
23. Holder WD jr WRj, Zuger JH. Effectiveness of positiron emission tomography for the detection of melanoma metastases. Ann Surg. 1998;227:764-769.
24. Swetter SM CL, Johnson DL, Segall GM. Positron emission tomography is superior to computed tomography for metastatic detection in melanoma patients. Annal Surg Oncol. 2002;9:646-653.
25. Tyler DS OM, Kherani A Positron emission tomography scanning in malignant melanoma. Cancer. 2000;89:1019-1025.
26. Brown RE, Stromberg AJ, Hagendoorn LJ, Hulsewede DY, Ross MI, Noyes RD, et al. Surveillance after surgical treatment of melanoma: futility of routine chest radiography. Surgery. 2010;148(4):711-6; discussion 716-7.
27. Morton RL CJ, Thompson JF. The role of surveillance chest Xrays in the follow up of high risk melanoma patients. Ann Surg Oncol. 2009;16(3):571-7.
28. Gupta SG, Wang LC, Penas PF, Gellenthin M, Lee SJ, Nghiem P. Sentinel lymph node biopsy for evaluation and treatment of patients with Merkel cell carcinoma: The Dana-Farber experience and meta-analysis of the literature. Arch Dermatol. 2006;142(6):685-90.
29. Warner RE, Quinn MJ, Hruby G, Scolyer RA, Uren RF, Thompson JF. Management of Merkel cell carcinoma: the roles of lymphoscintigraphy, sentinel lymph node biopsy and adjuvant radiotherapy. Ann Surg Oncol. 2008;15(9):2509-18.
30. Howle JR, Hughes TM, Gebski V, Veness MJ. Merkel cell carcinoma: An Australian perspective and the importance of addressing the regional lymph nodes in clinically node-negative patients. J Am Acad Dermatol. 2012;67(1):33-40.
31. Colgan MB, Tarantola TI, Weaver AL, Wiseman GA, Roenigk RK, Brewer JD, et al. The predictive value of imaging studies in evaluating regional lymph node involvement in Merkel cell carcinoma. J Am Acad Dermatol. 2012 Apr 30. [Epud ahead of print]
32. Concannon R, Larcos GS, Veness M. The impact of (18)F-FDG PET-CT scanning for staging and management of Merkel cell carcinoma: results from Westmead Hospital, Sydney, Australia. J Am Acad Dermatol. 2010;62(1):76-84.
33. Lu Y, Fleming SE, Fields RC, Coit DG, Carrasquillo JA. Comparison of 18F-FDG PET/CT and 111In Pentetreotide Scan for Detection of Merkel Cell Carcinoma. Clin Nucl Med. 2012;37(8):759-62.
34. Hong H, Sun J, Cai W. Anatomical and molecular imaging of skin cancer. Clin Cosmet Investig Dermatol. 2008;1:1-17.
35. Kim RH, Armstrong AW. Nonmelanoma skin cancer. Dermatol Clin. 2012; 30(1):125-39, ix.
36. Vidimos AT, Stultz TW. Imaging in cutaneous oncology: radiology for dermies. Dermatol Clin 2011;29(2):243-60, ix.
37. Weber RS. Improving the quality of head and neck cancer care. Arch Otolaryngol Head Neck Surg. 2007;133(12):1188-92.
38. Nemzek WR, Hecht S, Gandour-Edwards R, Ronald P, McKennan K. Perineural spread of head and neck tumors: how accurate is MR imaging? AJNR Am J Neuroradiol. 1998;19(4):701-6.
39. Vossen JA BM, Kamel IR Assessment of tumour response on MR imaging after locoregional therapy. . Tech Vasc Interv Radiol. 2006;9:125-32.
40. Boswell JS, Flam MS, Tashjian DN, Tschang TP. Basal cell carcinoma metastatic to cervical lymph nodes and lungs. Dermatol Online J. 2006;12(6):9.
41. Stella GM, Senetta R, Cassenti A, Ronco M, Cassoni P. Cancers of unknown primary origin: current perspectives and future therapeutic strategies. J Transl Med. 2012;10:12.
42. Kwee TC, Kwee RM. Combined FDG-PET/CT for the detection of unknown primary tumors: systematic review and meta-analysis. Eur Radiol. 2009;19(3):731-44.
43. Gourin CG, Watts T, Williams HT, Patel VS, Bilodeau PA, Coleman TA. Identification of distant metastases with PET-CT in patients with suspected recurrent head and neck cancer. Laryngoscope. 2009;119(4):703-6.
44. Kao J, Vu HL, Genden EM, Mocherla B, Park EE, Packer S, et al. The diagnostic and prognostic utility of positron emission tomography/computed tomography-based follow-up after radiotherapy for head and neck cancer. Cancer. 2009;115(19):4586-94.
45. Brix G, Lechel U, Glatting G, Ziegler SI, Munzhing W, Muller SP, et al. Radiation exposure of patients undergoing whole-body dual-modality 18F-FDG PET/CT examinations. J Nucl Med. 2005;46(4):608-13.