Guidance
6 Considerations
6.1 The Diagnostics Advisory Committee reviewed the evidence available on the clinical and cost effectiveness of using the VivaScope 1500 and 3000 imaging systems, to help decide whether to biopsy and excise skin lesions in people with suspected skin cancer, and to define the margins of skin lesions for excision in people with skin cancer, compared with current practice.
6.2 The Committee considered the quality of the studies included in the systematic review of clinical effectiveness. It noted that the External Assessment Group generally considered the studies to be at unclear risk of bias because insufficient information was reported in the publications. The Committee also discussed how clinical practice had advanced and heard from a clinical expert that older studies may not be representative of current NHS clinical practice. The Committee noted that more recent studies, from 2013 onwards, included care that was more representative of current clinical practice, and considered the impact of introducing confocal microscopy on clinical workflow. The Committee concluded therefore that studies from 2013 onwards were most relevant to the assessment: Alarcon et al. (2014), Pellacani et al. (2014), Ferrari et al. (2014), Castro et al. (2014), Stanganelli et al. (2014) and Rao et al. (2013).
6.3 The Committee considered the evidence on using the VivaScope systems to image different types of lesion. The Committee noted that there was a lack of available evidence on using the VivaScope systems in diagnosing lentigo maligna (LM) and in defining lesion margins in melanoma. The Committee heard from clinical experts that VivaScope is not useful in clinical practice for defining lesion margins in melanoma because the margins of melanomas are clearly defined and can easily be completely excised. The Committee also noted that no evidence was available on imaging squamous cell carcinoma (SCC) and recognised that improving diagnosis of this cancer is important. The Committee heard from clinical experts that SCCs can be difficult to view using imaging techniques because the upper surface is often scaly, which can make it hard to get sufficient penetration of the beam for effective imaging. The Committee also heard that confocal microscopes, including the VivaScope systems, do not currently have the technical capability to measure to a depth sufficient for accurately diagnosing invasive SCC. However, confocal microscopes may improve the diagnosis of in situ SCC if the carcinoma cells are confined to the epidermis and have not invaded the deeper dermis. The Committee concluded that the VivaScope 1500 and 3000 systems were not technically suitable for imaging invasive SCC and therefore, further research was not appropriate.
6.4 The Committee discussed the different types of biopsy used in the diagnosis and treatment of skin cancer. It heard that a punch biopsy is used in the diagnosis of skin cancers before treatment and that this type of biopsy is most commonly used for basal cell carcinoma (BCC) and LM. The Committee also heard that an excision biopsy is often used for melanoma and it can be used to diagnose and treat skin cancers simultaneously, or that it can be performed after diagnosis just for treatment purposes. The Committee concluded that it is important to understand the different types of biopsy used for different skin lesions to fully understand the different clinical pathways.
Diagnosis
6.5 The Committee considered the evidence on using the VivaScope systems after dermoscopy, to inform decisions on biopsy and excision of equivocal skin lesions in people with suspected melanoma and in people with suspected BCC. The Committee noted that the 2 studies (Alarcon et al. 2014, see section 5.12; Pellacani et al. 2014, see section 5.13) considered most representative of NHS clinical practice for melanoma diagnosis reported similar sensitivity values, but higher specificity values for the VivaScope systems compared with dermoscopy alone. However, the Committee also noted that the reported specificity values differed substantially between the 2 studies. The Committee considered the 1 representative study (Castro et al. 2014, see section 5.26) for BCC diagnosis and noted that although the reported sensitivity and specificity values were good, the study was small and at risk of bias because of patient recruitment and the lack of independent reviews of images. The Committee concluded that the evidence suggested that imaging using the VivaScope systems after dermoscopy had a higher negative predictive value than dermoscopy alone, but there is uncertainty in the actual accuracy values, particularly for BCC.
6.6 The Committee discussed the findings of a systematic review by Stevenson et al. (2013) that reported lower sensitivity and specificity values than those identified in this assessment. It heard from the External Assessment Group that the systematic review had been excluded from this assessment because it considered the accuracy of using the VivaScope systems in all people with suspected melanoma not just those with equivocal lesions post dermoscopy, as in this assessment. The Committee noted that this difference in population could explain the higher specificity values in this assessment. It also heard from the External Assessment Group that there is uncertainty in the accuracy values in the systematic review because the meta‑analysis combined patient‑ and lesion‑level data. The Committee concluded that it was not appropriate to include the systematic review in this assessment.
6.7 The Committee discussed the importance of training and experience in using the VivaScope systems and the impact on clinical effectiveness. The Committee heard from clinical experts that diagnosing skin cancer is dependent on experience and that there is considerable variation in the accuracy of diagnosis using current techniques. The Committee concluded that training on using the VivaScope systems in settings with sufficient numbers of skin lesions to ensure competency would be vital to achieving the higher negative predictive values reported in the studies compared with dermoscopy alone.
6.8 The Committee discussed the prevalence of BCC in people with positive or equivocal lesions. The Committee noted that in this group the prevalence was approximately 95% and the pre‑test probability was therefore high. It discussed the value of an additional diagnostic technology, such as the VivaScope system, in this population and noted that there was likely to be limited benefit and marginal improvement in accuracy. The Committee heard from clinical experts that for a person with BCC to begin treatment, a diagnostic biopsy is normally needed to confirm the diagnosis. The Committee concluded therefore, that using the VivaScope system may offer benefit in people with BCC by avoiding the need for a diagnostic biopsy, although the number of biopsies that would be avoided in clinical practice is uncertain.
6.9 The Committee considered the utility values used in the cost‑effectiveness model in diagnosing melanoma and BCC. It heard from the External Assessment Group that there were limited data available regarding utility values associated with anxiety from waiting for results, scarring from removing a lesion and getting a false positive result. The Committee noted that the utility loss associated with a skin biopsy or excision was lower than the disutility associated with the anxiety from waiting for results. The Committee discussed the plausibility of the size of the difference between these 2 values and noted there was considerable uncertainty around the utility values. The Committee noted that changes in these utility values could substantially affect the cost effectiveness of the VivaScope systems, which consequently results in substantial uncertainty in the ICERs.
6.10 The Committee considered the cost effectiveness of VivaScope in diagnosing melanoma and BCC in people with equivocal skin lesions. The Committee discussed the evidence and noted that there is uncertainty in the specificity of the VivaScope systems in diagnosing melanoma and in the accuracy of diagnosing BCC. It also noted the uncertainty in the number of biopsies that could be avoided and in the utility values used in the model. Overall, the Committee concluded that although the VivaScope systems show promise, there is too much uncertainty in the evidence for it to be confident that using the VivaScope systems represents a cost‑effective use of NHS resources.
Margin delineation
6.11 The Committee considered the clinical evidence for using the VivaScope systems to delineate margins of LM and noted that only 1 study had been identified and it had small patient numbers. It heard from clinical experts that lower recurrence rates could be inferred from the study but noted that the study was not comparative and had short (6 months) follow‑up, which limits the robustness of the findings. The Committee concluded that the VivaScope systems showed promise but further research was needed to determine their clinical effectiveness in defining margins of LM.
6.12 The Committee discussed the cost effectiveness of using the VivaScope systems to map margins of LM. The Committee noted that the ICERs suggest that using the VivaScope systems is cost effective (see section 5.69). However, it also considered the evidence informing the model and noted that there is substantial uncertainty in the diagnostic accuracy of the VivaScope systems and in the impact that their use has on lesion recurrence rates. The Committee concluded therefore, that there is too much uncertainty in the clinical evidence to determine if using the VivaScope systems is a cost‑effective use of NHS resources.
6.13 The Committee considered potential implementation issues of using the VivaScope systems to map the margins of LM. It heard from clinical experts that it took about 1 hour to map margins of LM using the VivaScope systems and the Committee noted that this was time consuming. However, clinical experts also informed the Committee that LM often needs very complex management and that spending time on accurate mapping can make treatment decisions more efficient, so the time spent mapping can effectively be offset. The Committee heard that people having surgery to remove a LM lesion may need to have up to 7 surgeries under local anaesthetic, spread over a number of weeks. This can result in people having open wounds between surgeries, which are at risk of infection. The Committee concluded that more accurate pre‑surgical mapping using the VivaScope systems could offer substantial benefits to people by reducing the number of surgeries.
General considerations
6.14 The Committee considered the training needed to accurately interpret the images produced by the VivaScope systems. It heard from the company that it currently provides training in Italy but is considering setting up a training site in the UK. The Committee heard from clinical experts that dermatopathologists are more familiar with interpreting cellular images than dermatologists and therefore, working together as a team may greatly help in developing the necessary skills and experience to interpret the images. The Committee concluded that effective training was vital to the clinical effectiveness of the VivaScope systems and was encouraged that training in the UK was being considered. The Committee considered the quality control of using the VivaScope systems in the NHS. It heard from clinical experts that there are currently no official quality control measures in place because the VivaScope systems are only being used in 1 hospital trust in England. The company informed the Committee that a Europe‑wide network of VivaScope users is being set up and that this would include quality control. The Committee concluded that a quality control scheme would need to be established to support widespread use of the VivaScope systems in the NHS.
6.15 The Committee discussed the numbers of people who would be examined using VivaScope for melanoma, BCC and LM. It noted that the greatest number of people would be examined for BCC, and that when VivaScope use included BCC the cost of the system spread across each lesion examined was greatly reduced. It also noted that the cost effectiveness of VivaScope was sensitive to the annual volume of suspected melanomas examined, if VivaScope was used exclusively for this purpose. It also heard from clinical experts that there is a lack of good quality data for the numbers of people following the different clinical pathways for melanoma, BCC and LM. The Committee concluded that there was uncertainty in the number of people who would be examined using VivaScope for the different skin lesions and consequent uncertainty in the number of biopsies and excisions avoided.
6.16 The Committee considered the impact of skin cancer on people. It heard from a patient expert that people can experience substantial anxiety about scarring and invasive procedures, particularly on the face and neck. People can also be shocked when first seeing a wound on their face, and the consequent scarring can lead to low self-esteem and withdrawal from social activities. The patient expert also highlighted that because skin cancer can be fatal in some people, the anxiety associated with biopsies, excision and the risk of skin cancer from their moles can be substantial and long‑term. The Committee also heard that anxiety can be even greater in people who have many moles. The Committee noted the points highlighted by the patient expert and acknowledged the substantial impact that skin cancer has on the lives of patients and their families.
6.17 The Committee considered the innovative nature of the VivaScope 1500 and 3000 systems and noted that the ability to provide images at a cellular level (quasi‑histological) in real‑time and in a near‑patient setting could offer substantial benefits to clinical practice. The Committee heard from a clinical expert that the technologies were promising and may lead to fewer biopsies, which would reduce the burden on pathology laboratories. The Committee also noted that pathology expertise is vital to interpreting images produced by the VivaScope systems and thought that the use of these technologies could encourage multidisciplinary decision‑making and the sharing of expertise, potentially improving the efficiency of the patient pathway. The Committee concluded that the VivaScope systems show promise but further research is needed to determine whether they are a cost‑effective use of NHS resources.
Research considerations
6.18 The Committee considered the lack of evidence on the disutility of anxiety, skin biopsy and scarring from excisions, and noted that it is unlikely that current estimates fully capture the disutility. The Committee encouraged further research on the disutility of skin biopsies and excisions and the associated anxiety.
6.19 The Committee heard from clinical experts that the VivaScope imaging systems may have potential for monitoring the incomplete response rate of topical chemotherapy and photodynamic therapy for treating BCCs. The Committee encouraged further research in this area.