Will the new Melafind device set a new standard in the early detection of Melanoma? 


JULY 2012 - It comes to no surprise to many in the dermatologist industry that the incidence of skin cancer is growing at an alarming rate in the US. Countless media headlines, hours of reaffirming scientific research papers and the continued targeting of billions of dollars in the R&D departments of Goliath pharmaceutical  companies on the treatment of such aliments; reinforce the scope of this disturbing global trend. 

Until very recently, melanoma; the deadliest of all occurring skin cancers had very little effective or new late stage treatment options. In 2002, an important research paper changed this. In the top science journal Nature, called "Mutations of the BRAF gene in human cancer" 

Researchers discovered that a gene called BRAF was faulty in about 7 out of 10 cases of melanoma. Isolating a particular fault in this gene (V600E) was found responsible for 5/10 melanoma cases. 

After this important discovery, Vemurafenib or Zelboraf was launched targeting this mutation. With other new drugs currently in development. 

It is important to note that these promising new drugs are not for everyone. As a rough estimate, take the UK as an example - with 11,000 people diagnosed per year with melanoma, approximately 500-1000 of these people could show a positive response to this new drug. Also important to note, this is not a cure. The cancer often overcomes the drug in a relatively short time, it is extremely expensive, can have serious side effects and is not yet widely available. 

Simply stated, the best cure for skin cancer of all types remains early detection. It offers the patient the highest chances for a cure and the cheapest, most effective option for treatment. Saying this, especially for melanoma, timing is so very critical. For melanoma in situ, the 5 year survival rate is virtually 100%, however metastatic melanoma it is less then 20%. Catching these hard to diagnose early cancers is a challenge for many. 

The current eyeballing method that Dermatologists and Medical Doctors use in the field is subjective at best. Little technologic advances have been developed to help augment this visual subjectivity - that is until now. There are two devices that have caught the attention of those knowing the risks, Mela Sciences Melafind using multispectral imaging and Verisante Technology's Aura using rapid Raman spectroscopy. 

Mela Sciences Melafind (above) 

Verisante Technology's Aura (above) 


Firstly, Melafind has been approved for use by trained dermatologists in the US by a narrow FDA vote of 8-7 in favor with one abstaining. Verisante Technology's Aura is expected to be FDA approved by latest 2014 however it has received approval for sales in Canada, Australia and the European Union expected to commence later this year. 

MelaFind is the first FDA approved computer aided device for melanoma detection. According to the development team, "MelaFind is a novel multispectral digital dermoscope, with "expert system" software that performs the image processing sequence automatically end-to-end, leading to differentiation between early melanoma and benign atypical neoplasms. This digital dermoscope captures image data both in the visible and in the infrared (a spectrum that lies outside the eye's sensitivity). Image acquisition is done under computer control to ensure standardization. Next, the digital dermoscope uses the image data to uncover new features that can only be obtained with computer assistance. Finally, it uses the potent mathematical framework of statistical pattern recognition to accomplish differential diagnosis."

The journal, the Archives of Dermatology published a prospective, multicenter, blinded study on the diagnostic performance of MelaFind vs dermatologists, which was evaluated using the histologic reference standard. In this trial, MelaFind achieved a high sensitivity to thin melanomas and borderline lesions of >98%.  The average specificity was 9.5% Both of these parameters were better than the clinical diagnoses made by the investigators.

Great news! However, Melafind may have some key drawbacks to consider:


MelaFind is designed to be an aid in evaluating pigmented lesions that have one or more clinical or historical characteristics of melanoma.  That means the clinician has to identify the lesion as suspicious and then examine it with the MelaFind. According to the agreed-upon labeling, "MelaFind is designed to be used when a dermatologist chooses to obtain additional information for a decision to biopsy. MelaFind should NOT be used to confirm a clinical diagnosis of melanoma." 

The device is indicated only for use on lesions that meet a number of specific criteria, including:

          * Accessibility by MelaFind tool (very large probe head)

          * Diameter of 2 to 22 mm

          * Sufficient pigmentation

          * No scarring or fibrosis

          * Non-ulcerous and non-bleeding

          * Not on acral, palmar, plantar, mucosal, or subungual areas

          * Located >1 cm from eyes

Healthcare professionals must also complete a training program in the proper use of the device before using MelaFind.

Statistical method used to calculate the device sensitivity at 98% was chosen by Mela Sciences as a one-sided exact Mid-p binomial method. Some view this as a weaker statistical evaluation. The FDA certainly voiced this concern despite approving. Comparing this method to the Clopper-Pearson, Score methods and the two-sided Mid-p method the FDA showed much lower results.  

Please see: http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/MedicalDevices/MedicalDevicesAdvisoryCommittee/GeneralandPlasticSurgeryDevicesPanel/UCM233854.pdf



Most importantly in my mind, is the issue of Melafind Non-Evaluables- these were not included in the assessment of melafind's accuracy but should be taken into consideration. Let's put this into perspective: in Melafind's clinical study, of the 1831 enrolled lesions, 162 were deemed to be non-evaluable, meaning scanned with no result given, of which included 27 pathology confirmed melanomas. Once hair is shaved, alcohol applied and suspicious lesion scanned using the Melafind device, approximately 9% of all these scanned lesions were non-evaluable by the device and of those lesions, 16.7% were confirmed melanomas. 

Some may say - 'Heck FDA –don't be so  tough on the new kid on the block, Melafind is the first of its' class, you need to cut them a break and let this new device out there so it can maybe get better through it's use and be refined with next generation development' - indeed with Mela Sciences successful citizen petition campaign – this is what happened. This will certainly be a good thing for Canada's Verisante Technology. 

Melafind has successfully carved out the path towards FDA approval for Verisante's AURA. This is the next generation of early cancer detection today. 

The Verisante Aura was developed through over 10 years of dedicated work through Canada's British Columbia Cancer Agency and The University of British Columbia's Department of Dermatology. 

For those unfamiliar, please see this very descriptive video outlining Verisante Technology and it's award winning Raman spectroscopy device platform for the early detection of skin, lung, colon and cervical cancers. 

Quickly broken down, here are the advantages that the Verisante Aura shows over it's closest competitor Melafind: 

  • completely non-invasive system uses Raman laser light to quickly (under a second) analyze a suspicious lesion through scanning 21 different cancer biomarkers at the cellular level. 
  • when using this Raman system healthy cells and diseased cells  were found to reflect differently, it's like the healthy cells sing a different song than the unhealthy tissue when interpreted by the Aura device, and all this happens in under a second. 
  • quick scan time offers the possibility for full body scanning of multiple suspicious lesions in a single patient appointment.
  • Aura detects melanoma at a higher accuracy then the Melafind – 99% sensitivity and 14% specificity using the very conservative statistical method of PLS (partial least squares) versus Melafind's 98% sensitivity and 9.5% specificity using one-sided mid-p method (discussed above). This leaves many wondering what Verisante Aura's results would have looked like if their study used the same one-sided mid-p statistical analysis. 
  • Aura when used received 100% results for 100% of scanned lesions. Melafind when used received 91% results for 100% of scanned lesions – remember in Melafind's clinical study: of the 162/1831 scanned lesions that returned a non-evaluable result, pathology after biopsy revealed 27/162 as melanoma. 
  • Aura can detect with high accuracy not only melanoma but also squamous cell and basal cell carcinomas – two very common types of non-melanoma skin cancers, and actinic keratosis, a premalignant skin condition – helping achieve the earliest possible patient treatment for skin cancers. Melafind is only indicated for detection of melanoma in highly atypical lesions. 
  • Many skin cancers occur on the hard to reach spots on the body such as around the ears and other narrow hard to see areas. A narrow probe head with a quick scan time is essential for a timely diagnosis. 

The Aura's probe head (below) with a scan and result in under one second 100% of the time: 

Melafind's probe head (below) with a scan and a result in under 1 minute 91% of the time:


The jury is still officially out on this new technology. Melafind certainly has a first to market advantage in the USA however Canada's Verisante Technology and the Aura shouldn't be too far behind. 

The best outcome either way is for current and future skin cancer patients. Early cancer detection saves lives period – now we just need the new tools that will allow this to happen more often! 


Disclosure: These are just my own personal thoughts and interpretations. Do your own due diligence before any investment in the firms mentioned. 

Rate this post
0 stars