JD v Mather

This is the trial of the issue of causation in a clinical negligence claim. The Claimant consulted the Defendant, his general practitioner at her surgery on 8^th March 2006 because he was concerned about a growth in his right groin. That consultation has already been the subject of a fact finding hearing before Mackay J, whose findings are summarised in the next paragraph of this judgment.

In his reserved judgment dated 8^th July 2011 Mackay J found that the Claimant had registered the presence of the growth by mid-2005, and had been worried about it for some months before going to the surgery in March 2006. He went to see the doctor because the lesion “appeared to him to be growing, it was itching, and it had bled when he had scratched it”, and mentioned all three features to the Defendant.

At the surgery the Defendant looked at the area of the lesion closely but briefly and formed the opinion that it was a seborrhoeic wart, which she described in her witness statement as having a typical “stuck on appearance and demarcated borders”. She reassured the Claimant that he had nothing to worry about. She suggested that the fact that it had bled was probably due to it having been caught when he was zipping up his trousers. She appears to have been more concerned about two lesions on his chest, which turned out to be benign. Unfortunately Dr Mather was entirely wrong in her diagnosis of the groin lesion.

On 16^th October 2006 the Claimant visited the surgery again. The Claimant’s wife, who is a nurse, looked at the lesion before he went to see the doctor. Her evidence was that it was “very dark, dry and quite big”.

This time the Claimant was seen by the Defendant’s father, Dr Andrew Mather. He found that by this time the lesion was “huge” with a surface area of 15 x 20mm and realised that it was possibly a malignant melanoma. He removed it by curettage and sent it to Stepping Hill Hospital, Stockport for histological examination. It was examined there by Dr Asma Naveed. Her macroscopic findings were as follows:-

“Two portions of skin. The smaller is 0.9 x 0.7 x 0.3cm. The skin surface is irregular and contains a hyper-pigmented pigmented lesion which occupies most of the skin’s surface. The other piece of skin is 1.5 x 1.0 x 0.5cm. It shows a smooth irregular hyper-pigmented lesion which covers the skin’s surface.”

Dr Naveed’s microscopic examination revealed the following:-

“This is ulcerated skin containing a nodular vertical growth phase malignant melanoma. … The surface is extensively ulcerated and, whilst the exact size of ulceration is difficult to measure due to the nature of the specimen, it measures at least 12mm. The tumour shows a brisk mitotic activity (34 mitoses per 10 high power fields). There is no evidence of lymphovascular or perineural infiltration. The Breslow thickness is difficult to assess accurately due to poor orientation in place, however, in better orientated fragments it measures 5mm and the Clark Level is 4 in this specimen although no subcutaneous fat is included. There is associated (sic) with moderate active and chronic inflammatory cell response. There is no evidence of regression satellite nodules or pre-existent naevus. Due to the fragmented nature of the specimen, completeness of excision cannot be accurately commented on, but it likely to be incomplete.”

The Breslow thickness of a tumour is the vertical distance from the surface of the skin to the lowest cancerous cell. Dr Naveed concluded that this was a nodular malignant melanoma in the vertical growth phase with a Breslow thickness of 5mm.

Dr Naveed sent a report containing these findings to Dr Andrew Mather. He in turn made an urgent referral to the Stepping Hill Hospital, by fax dated 23 October, on a form issued by them which asks the referring doctor to tick a series of boxes “yes” or “no”. He ticked four “yes” boxes to indicate that the melanoma was growing in size, changing shape, changing colour and inflamed. There was also a box for the feature “oozing/ulceration” to which he answered No. He also faxed with the referral form a copy of Dr Naveed’s findings.

On 27^th October 2006 the Claimant was seen by Dr Gillian Muston, an associate specialist in dermatology at Stepping Hill. She had the unenviable task of explaining the histopathology report to the Claimant. She referred him to Mrs Brain, a consultant plastic surgeon at the Christie Hospital, which is a specialist referral hospital in this field. He was seen there 3 days later. It was found that the curettage had left a remnant of the melanoma in the Claimant’s abdomen. On 14^th November 2006 this was excised. A histopathological report from the Christie dated 30^th November 2006 reads as follows:-

“Macro: A skin ellipse measuring 65mm x 30mm x 30mm, which shows a scarred area on the surface measuring 8mm x 5mm (with focal dark pigmentation). Attached to the skin is a specimen of groin dissection measuring 105mm x 90mm x 30mm. 8 lymph nodes are present in the fat…

Micro: Sections from the skin show a focus of residual in situ and invasive malignant melanoma. The maximum horizontal extent of this focus is 5mm. The Breslow thickness in this excision is 2.8mm (Clark’s level IV). (This may not be accurate due to previous excision). There is no evidence of ulceration… two out of eight lymph nodes identified contain metastatic melanoma.”

In early 2007 the Claimant wrote to the practice to complain of the mis-diagnosis. The first letter in response confused the chest lesions with the groin lesion, in particular by saying that the lesion had been “with a flat surface (that is to say not raised or ulcerated), and asymptomatic (that is to say with no altered sensation, itch or bleeding noted)”. The Claimant challenged this in a reply dated 14^th February 2007, saying: “my own memory was that the mole was raised above the surface of the skin and was ulcerated – i.e. had bled”. In a later letter to the Healthcare Commission he repeated that “my memory is that the mole had bled”. On 19^th July 2007, in response to a letter from the Commission, the Defendant wrote that her usual practice when assessing pigmented lesions had been “to use an approach recognised by the British Society of Dermatologists, that is to assess for change in size, assess for irregular shape, irregular pigmentation, and also assess diameter and for symptoms of inflammation/oozing.” She added that dermatology had been an area of specialist interest within her GP practice.

In the light of Mackay J’s findings the Defendant’s solicitors, by letter of 4^th November 2011, admitted that she had been in breach of her duty of care on 8^th March 2006, but made no admission as to the nature or extent of any causative damage flowing from that negligence.

It is common ground that if the Defendant had referred the Claimant to a specialist with a suspected malignant melanoma in March 2006, as she should have done, he would have been seen within two weeks and the tumour would have been excised. The Claimant’s case as originally put in the Particulars of Claim was that “excision and treatment at this time would have carried a likelihood of cure and survival”. A fallback case, not spelt out until Professor James wrote a supplementary report very shortly before the trial, is that excision and treatment in March 2006 would at least have given the Claimant a longer period of disease-free survival. Sadly the cancer is now classified as Stage IV: it has reached his lungs, and the prognosis is very poor.

I have before me the witness statements of the Claimant, his wife and the Defendant, all of whom gave evidence before Mackay J; there was no application to call or cross-examine any of them at this hearing. No oral evidence or witness statement has been adduced from Dr Andrew Mather, Dr Naveed or the author of the Christie report of 30^th November 2006. The oral evidence before me was given by three pairs of experts: histopathologists (Dr Robert Vanhegan for the Claimant and Dr Andrew Boon for the Defendant); dermatologists (Dr Iain Foulds for the Claimant and Dr Jeremy Marsden for the Defendant); and clinical oncologists (Professor Roger James for the Claimant and Dr Paul Nathan for the Defendant).

The prognosis for a patient with a malignant melanoma depends on whether the cancer has spread into either the blood stream or the lymphatic system. Once it has, the cancer is likely to move via the lymph nodes to the vital organs, where sooner or later it will prove fatal. The nearest lymph node to the tumour is generally referred to as the sentinel.

The American Joint Committee on Cancer (AJCC) staging system, to which all the expert witnesses before me referred, uses four predictive factors: (a) the Breslow thickness of the primary tumour; (b) whether the primary tumour is ulcerated; (c) whether the cancer has reached one or more lymph nodes, and if so whether it is palpable or only detectible under a microscope; (d) whether there has been metastatic spread. The last of these was not discussed in the evidence before me.

In his report of 2 February 2012 Dr Vanhegan wrote that the Breslow thickness of the primary tumour in the autumn of 2006 was 2.8mm (the figure given in the Christie report of 30 November), and calculated that in March 2006 it would have been 1.6mm. Dr Foulds and Prof James, in their reports, fell in behind Dr Vanhegan in using the figure of 2.8mm. But this is fallacious. The Christie report was not on the tumour which Dr Andrew Mather had curetted on 16 October and which Dr Naveed had examined at Stepping Hill the following week; it was on the residual tumour which was excised on 14 November. This was likely to be only a small proportion of the original tumour, and the measurement of 2.8mm is therefore of little if any significance. As Dr Vanhegan eventually conceded in cross-examination, the correct measurement to use for overall Breslow thickness in October 2006 must be no less than, and probably more than, Dr Naveed’s figure of 5mm for the curetted part of the tumour. It was unlikely to be as much as 7.8mm, since that would be to assume that (improbably) the curettage was carried out exactly parallel to the surface. Dr Boon’s evidence was that it can be estimated “with a reasonable degree of confidence as at least 5.5mm”; Dr Marsden’s figure was “at least 5 mm”. I do not think it matters which of them is right. The important issue is the thickness in March 2006. I accept Dr Marsden’s evidence, which in the end was scarcely disputed, that this is likely to have been in the bracket 3-4 mm.

There is a problem of terminology in this aspect of the case. An ulcerated tumour is one the surface of which is bleeding or oozing. It was common ground that if this is the result of an operation (artefactual ulceration), or of abrasion caused, for example, by the patient scratching the surface (traumatic ulceration), it is of no diagnostic significance; but if it is caused by the cancer (malignant or “true” ulceration), it is of considerable significance and indicates a reduced prospect of 10-year survival..

The Royal College of Pathologists has issued a document “Minimum Dataset for the Histopathological Reporting of Malignant Melanoma”. The section on ulceration states:

“The presence of ulceration, particularly when greater than 3 mm in diameter is an independent prognostic indicator associated with a diminished five-year survival rate. It should be noted that ulceration can ‘downstage’ the tumour thickness.

Ulceration has recently taken on an additional degree of significant importance. This follows new proposals from the New American Joint Committee on Cancer [AJCC] Staging System for Cutaneous Melanoma …... Melanoma ulceration is now regarded as the second most important prognostic variable and is independent of tumour thickness on multivariate analysis. It also enjoy a high degree of inter-observer agreement. Ulceration is defined here as the absence of intact epidermis overlying a portion of the primary melanoma, based on microscopic observation. Unfortunately, no distinction is made between biological ulceration and traumatic/exogenous ulceration. Likewise, a minimal width of ulceration is not stated.”

In the 2001 final version of a paper on the AJCC Staging System the authors (Charles Balch and others) wrote:-

“Melanoma ulceration is defined as the absence of an intact epidermis overlying a major portion of the primary melanoma based on microscopic examination of the histologic sections. It can easily be distinguished from artifactual or traumatic disruption of the epidermis. Traumatically induced defects are associated with haemorrhage, brightly eosinophilic fibrin exudation at the site, and an architectural defect that usually defines the agent leading to the trauma such as an insect bite or an excoriation. In fact, the interpretation of melanoma ulceration among pathologists is one of the most reproducible of all the major histopathologic features. This definition encompasses surface defects from a total absence of the epidermis overlying the tumour to an excavated area including the epidermis and a portion of the tumour. The surface may exhibit scattered debris.”

However, in their book Pathology of Melanocytic Disorders, published in 2007, Mooi and Krausz wrote of this passage:-

“The AJCC guidelines state that ulcerations can easily be distinguished from artifactual or traumatic disruption of the epidermis. In our experience and that of others, this is incorrect. We regard this distinction possible only with regard to disruptions occurring during or after removal of the melanoma, not to previous trauma. We do not agree that “an architectural defect usually defines the agent leading to the trauma, such as an insect bite or an excoriation”. In fact, we do not have experience with melanoma complicated by an insect bite; this must be an extraordinary occurrence.

In accordance with Spatz and colleagues, we consider ulceration to be present in any situation where there is a full-thickness defect of the epidermis covering the melanoma, the defect is covered by fibrin and cell debris with neutrophils and there is associated reactive change (thinning, effacement or reactive hyperplasia) of the immediately adjacent epidermis. These features distinguish true ulceration from artifactual damage during or after surgical removal and assessment of presence of ulceration based on this description was much improved when compared to its assessment according to the less precise AJCC guidelines.”

Ms Hatfield relied upon: (a) the absence of any complaint by the Claimant about oozing; (b) the evidence of the Claimant’s wife that when she saw the lesion in October 2006 it was dry; (c) Dr Andrew Mather’s finding that on 16^th October 2006 the tumour was not showing oozing/ulceration. As against this Mr Kennedy relied on the Claimant’s use of the word “ulcerated” in his complaint; the finding of Dr Naveed that the tumour was ulcerated; and the subsequent classification of the Claimant’s case by the specialist team at the Christie as Stage IIIC which implies an acceptance by them that the tumour had been ulcerated. Importantly, Mr Kennedy also relied on enlarged photographs of the curetted specimens of the tumour produced by Dr Boon shortly before the trial. Two of these in particular, which became pages 153B and 153C of the trial bundle, became the heart of the Defendant’s case on ulceration, since Dr Boon and Dr Marsden both consider that they clearly show malignant ulceration. Page 153B clearly shows, even to the eye of a layman, that on the left of the picture the skin surface is largely worn away, whereas on the right of the picture it is intact but thinned. Beyond that I had to look to the experts for interpretation.

To get one point out of the way: I attach no significance to the Claimant’s use of the word “ulcerated” in his letter of 14 February 2007, especially since he went on to write “ie had bled”, and was responding to a confusing letter from the Defendant’s practice.

Dr Vanhegan considers that a mechanical or traumatic ulcer tends to have a sharp edge, whereas a malignant one does not; and that the picture does show a sharp delineation between the ulcerated and non-ulcerated sections. He also considers it unlikely that malignant ulceration would cover only part of the surface of the tumour, whereas it is obvious that traumatic or artefactual abrasion can occur over part of the surface. He considers that the epidermis on the right of page 153B, though “pushed up” by the tumour and thinned, is not invaded by the tumour as it would be if it were affected by malignant ulceration. In his view the cause of the apparent ulceration of the surface of the Stepping Hill specimen was abrasion at the start of the curettage. However, he conceded in cross-examination that a slide showing a true malignant ulceration would “not necessarily differ” in appearance from page 153b. He placed emphasis on the clinical history. He accepted that there was a full thickness defect of the epidermis and that this was covered by fibrin, neutrophils and cell debris, but did not see any reactive change of the immediately adjacent epidermis.

Dr Boon had no doubt that the slides enlarged at pages 153A-E show true melanoma ulceration; in his view it was “not open to doubt”. He considered that there was no sharp line on page 153B between the ulcerated and non-ulcerated areas. The ulceration, he said, overlies a major proportion of the primary melanoma. Dr Naveed’s finding that the ulcer was 12mm in diameter was important. His team at Leeds (where he has been the lead melanoma pathologist since 1996) regularly have debates about scratches on tumours but he considers that the photographs in this case show “tumour ulceration beyond a shadow of a doubt”. All the elements mentioned in the Mooi and Krausz book are shown: a full thickness defect of the epidermis covering the melanoma, the defect being covered by fibrin and cell debris with neutrophils and thinning of the immediately adjacent epidermis. He would not set page 153B as a test for a student taking examinations at the end of the first year of study at the Royal College of Pathologists because, it is too easy. He attached no significance to the fact that ulceration does not cover the entire surface of the tumour: it would, he said, be exceptional for this to occur. He agreed that small areas of malignant ulceration can be confused with abrasion from scratching; but a number of clinical trials have shown that ulceration of more than 3mm in diameter is malignant. As to the supposed hard edge between the ulcerated and non-ulcerated sections, he disagreed that the photographs show any such clear distinction. He was sure that with a few more days in situ the right hand part of the surface shown at page 153B would have become ulcerated as well. He emphasised the Royal College guideline that ulceration over 3mm in diameter is an independent indicator of a diminished survival rate.

Dr Boon also said – and this was scarcely in dispute by the end of the case - that little if any significance can be attached to the fact that the residual tumour excised at the Christie showed no evidence of ulceration. On any view Dr Andrew Mather had removed the original surface and the bulk of what was underneath. It is quite possible, according to Dr Boon, that the residual tumour was not ulcerated, or that it had not had enough time since the curettage to grow enough to ulcerate.

Dr Foulds, who had heard the evidence of Dr Boon on ulceration, told me that he was not convinced by it, although conceding that it was not his speciality. He drew attention to the lack of any history of oozing either from the patient or from Dr Andrew Mather’s referral form. He supported Dr Vanhegan’s view that the slides show a sharp cut-off between the ulcerated area and the rest. He said that he was “not persuaded one way or the other” as to the malignancy of the ulceration, and that it was “difficult to reach a conclusion”. But on balance he adhered to the view that if it was malignant he would expect it to have been oozing; he said that he could not see how a tumour could have been malignant with ulceration for seven months without evidence of oozing.

A final point made by Dr Foulds as to which there is no dispute is that, if a tumour is reported as ulcerated, those reading the report should take no chances on whether it is traumatic or malignant.

Dr Marsden, who is the lead author of the 2010 UK melanoma guidelines and the lead melanoma clinician at University Hospital Birmingham, supported Dr Boon. He regarded page 153B as showing “simple common or garden melanoma ulceration” which would have been present in March 2006, this being “by far the most frequent explanation”. He regarded this as a “typical ulcerated melanoma” with an absence of epidermis in the area of the ulcer and thinning of the adjacent epidermis. He said that patients often describe oozing as bleeding; and it was very unusual in his experience for a mole simply to bleed. Recurrent bleeding suggested an ulcerated surface which bleeds readily on contact. Here the patient already had a history of repeated bleeding when he went to the doctor in March 2006. The combination of this with the presence of a melanoma is strongly suggestive of ulceration.

Professor James was not put forward as an expert on ulcers, but said that his experience of true or malignant ulceration of tumours was that the patient is continually affected by suppuration of pus with smell and occasional bleeding. He could not say whether if malignant ulceration was present in November it would also have been present in March.

Dr Nathan, perhaps wisely, did not offer an opinion on this topic.

The clinical history does help the Claimant’s case on ulceration, but it is by no means conclusive. Mackay J accepted, as do I, that by the time the Claimant went to see Dr Melanie Mather in March 2006 he had had both itching in the area of the groin lesion and bleeding from it. However, he has not said that he only experienced bleeding when he had scratched the surface of the lesion; and nothing was said about oozing.

Thus far the clinical history is equally consistent with either side’s case. Ms Hatfield’s best point is the combination of the Claimant’s wife’s evidence that the tumour was dry when she saw it in October 2006 with Dr Andrew Mather’s reporting that it was not oozing or ulcerated. But in my judgment this is outweighed by the evidence of Dr Boon, supported by Dr Marsden, as to the conclusions to be drawn from the enlarged slides. It is possible that Dr Andrew Mather had damaged the surface in the process of curettage. But it seems improbable that if he had, the photographs of the surface and the underlying tumour should nonetheless exhibit so clearly the features of a typical malignantly ulcerated tumour. I attach weight to Dr Vanhegan’s acceptance that a slide showing malignant ulceration would “not necessarily differ” from what is shown at page 153B. I also attach weight to the fact that, while Dr Vanhegan retired from practice 10 years ago, Dr Boon (and likewise Dr Marsden) has for some years been at the head of a specialist melanoma team considering hundreds or even thousands of cases per year. I find that on the balance of probabilities the tumour was already the subject of malignant ulceration in March 2006.

Most patients whose melanomas are only at Stage II do not suffer a relapse (ie the spreading of the cancer into the lymphatic system or the bloodstream) within 10 years. Dr Marsden told me, and Dr Nathan confirmed, that out of every 10 such patients 2 suffer a relapse in the first two years from diagnosis, 2 after two years but before 10 years, leaving 6 who do not relapse within 10 years. We know that the Claimant’s cancer had reached his lymph nodes by October 2006. The issue is whether it is more probable than not that it had already done so by March of that year.

Dr Marsden, in his report of 27 April 2012, wrote as follows:

“Melanomas which are 3-4mm thick and ulcerated have about a 30% risk of developing involvement of the regional lymph nodes. This may be detected by sentinel lymph node biopsy if the nodes are clinically normal, or by clinical examination if the nodes are palpable. We do not know that [Mr D’s] lymph nodes were clinically normal in March 2006 since they were not examined. However, we do know that the regional lymph nodes were clinically abnormal and histologically positive for melanoma in October 2006. the probability of developing palpable regional lymph node involvement in a population of patients with at least 5.0 mm thick ulcerated melanomas is about 45%. However, [Mr D’s] risk of developing regional lymph node involvement in March 2006 when the melanoma was 3.0 mm thick was not simply 30%, as he has since been found to be one of the 45% who were destined to become positive. For the risk in March 2006, one needs to work back from the fact that we know the lymph node was involved 7 months later, and therefore the risk was not simply that of the general population at 30%. Evidence indicates that to arrive at a realistic figure the 30% should be adjusted to reflect the position actually found in October 2006. Therefore his actual risk when the melanoma was 3-4 mm thick was a value between population risk, ie 30%, and 30% divided by 45%, or 66%.

The average interval between diagnosis of primary melanoma and first clinically detectable relapse is 2 years. The interval between the first opportunity for diagnosis of [Mr D’s] primary melanoma and his first relapse was 7-8 months. Consequently, in March 2006, whether or not the nodes were clinically enlarged, they were very likely, on the balance of probabilities, to have been involved. My opinion is that in March 2006 [Mr D’s] melanoma was T3b/T4b N1-2a M0 (stage IIIB melanoma) or T3b/T4b N1b M0, (stage IIIC melanoma). This means that in March 2006 he had an ulcerated primary melanoma and either 1 or 2 involved but not enlarged, sub-clinical lymph nodes, or an ulcerated primary melanoma and 1 enlarged lymph node.”

The first of these paragraphs was put to me as an example of conditional probability theory. I am not a statistician any more than Dr Marsden is, but I am uneasy about his suggested range of 30-66%. I accept his evidence that 30% of melanoma patients with a tumour 3mm thick go on to suffer regional lymph node involvement (a figure confirmed by research papers of Rousseau and Lee adduced by the Claimant’s team), whereas 45% of those with a 5mm thickness do so. And we do know that by October 2006 the Claimant had developed palpable regional lymph node involvement. But this is not, as I see it, enough in itself to prove that his likelihood of doing so in March 2006 was two-thirds, or somewhere between 30% and two-thirds. There is a difficulty about applying such statistics to show an individual cancer patient’s prognosis for the purposes of a trial of causation: see Gregg v Scott [2005] 2 AC 176 at [153]; Sinkiewicz v Greif UK Ltd [2011] 2 AC 229 at [163].

As to Dr Marsden’s second paragraph, the proposition that the average time from diagnosis to the first clinically detectible relapse (among those who relapse at all) is two years is supported by a German survey based on several thousand patients. Mr Kennedy was also able to rely on the clinical experience of Dr Nathan, who told me that “only a tiny proportion of patients who have had the primary tumour resected relapse within a year”. This is confirmed by a study by Morton and others, published in the New England Journal of Medicine in September 2006, showing that of an observation group of 500 patients, 78 relapsed within 5 years; that among these 78 the median time to clinical detection of nodal relapse was 1.33 years; and, most significantly, that 95% of the 78 relapsed in the period 1.02 to 1.76 years from detection. This was a relatively small sample: a much larger study conducted in Germany by Leiter and others found that 14% of patients with Stage II melanomas relapsed within a year, but that is still a small minority. These research papers essentially corroborate Dr Nathan’s experience based on a large number of patients over many years.

Ms Hatfield suggested that all such studies were of limited applicability to the present case. They all concern patients whose primary tumour was detected – and thus, of course, promptly excised, since to do otherwise would be wholly unethical – before time began to run; not left in situ as the Claimant’s was. She argued that this made reliance on the Morton and Leiter statistics misconceived: supported by Professor James, who argues that studies by White at al and Fournier et al in 2002 and 2008 respectively show that patients who develop lymph node involvement with the melanoma still in place (synchronous cases) have significantly worse outcomes than those who develop such involvement after resection (metachronous cases, where the nodes are described as interim nodes).

However, this is not what the White and Fournier papers show. White’s researches demonstrate, perhaps counter-intuitively, that patients who develop metastases within one year after excision of the primary tumour had a slightly lower survival rate than those whose metastasis occurred with the primary still in place. Fournier’s definition of “synchronous” disease includes patients who develop lymph node involvement up to one year after excision of the primary. Their studies were directed at the usefulness or otherwise of Cyto-reductive surgery in the form of dissection of cancerous lymph nodes, as a means of prolonging the lives of patients whose melanoma has already metastasised as far as the regional lymph nodes.

Dr Marsden and Dr Nathan both told me that there is no evidence to support the proposition that where the primary tumour remains in place the speed of metastasis is increased. Dr Nathan said that the reason a primary tumour must be excised immediately on detection (and the reason for the much publicised two week referral time) is that if it has not already spread, it might do so at any moment. But once it has spread, he said, there is no evidence that excision of the primary tumour makes a difference: the metastasised cells, for whatever reason, acquire a life of their own. Dr Nathan also pointed out, as is obvious, that the cohorts studied must include many where the diagnosis and consequent excision had been delayed, in the sense that the tumour had been present for some time before the patient consulted a doctor. Even taking that factor into account, relapse in the first year after diagnosis is rare. The Claimant’s witnesses were unable to contradict this central point.

Ms Hatfield pointed out that neither the witnesses before me nor the research papers indicate that relapse within 8 months is impossible; that the risk judged prospectively in March 2006 was only 30%; and that therefore the Defendant has not rebutted the presumption in the Claimant’s favour created by the figure of 30%. But this argument appears to me to reverse the burden of proof.

I therefore accept the Defendant’s argument that it is rare for a tumour which has not spread at all to give rise to palpable lymph nodes within 7-8 months; and that it is far more likely than not that the Claimant’s tumour had spread with at least microscopic regional lymph node involvement by March 2006. I am not satisfied that the regional lymph node involvement was by then palpable or macroscopic.

The result is that on the balance of probabilities the Claimant’s melanoma in March 2006 was at Stage IIIB of the AJCC staging, namely an ulcerated primary tumour with at least one regional nodal micrometastasis. (By October 2006 it had reached Stage IIIC, an ulcerated tumour with two macroscopic regional nodes).

Counsel are agreed that if that was the position in March 2006 the Claimant’s chances of surviving a further 10 years were already less than 50% , even if the tumour had been detected and excised then as it should have been. His principal claim must therefore be dismissed, in accordance with the majority decision in Gregg v Scott.

Professor James, in a late addendum report, considered what would have been the effect of diagnosis and treatment in March 2006 on the basis of the court finding (as I have found) that by this time the tumour was ulcerated and that there was microscopic nodal spread. He wrote:

“In my view the Claimant would in these circumstances have had primary resection with either sentinel node biopsy, careful follow-up or entry into a clinical trial. Assuming a sentinel node biopsy were performed and microscopic melanoma were detected, the Claimant would have proceeded to lymph node dissection; assuming follow up rather than sentinel node biopsy, clinical regional nodes would have been detected at some point. In my opinion the net result of earlier treatment would have been extended life expectancy, depending on the impact of surgery on the “dormancy” of the metastatic deposits……In my opinion the extended life expectancy would have been of the order of a year.”

This alternative case was not in the original Particulars of Claim, and Professor James’ addendum report was only served just before the trial. (On the other hand, so were Dr Boon’s photographs 153A-E, which as will have been noted were of considerable importance to the Defendant’s case.) Mr Kennedy fairly and sensibly did not seek to exclude it, and I gave leave for the necessary amendment at the conclusion of the hearing. It is plainly a claim of a permissible type. As Baroness Hale said in Gregg v Scott at [206]-[207]:

“The defendant is liable for any extra pain, suffering, loss of amenity, financial loss and loss of expectation of life which may have resulted from the delay. If, without the delay, the claimant would have achieved a longer gap before more radical treatment became necessary, then he should be entitled to damages to reflect the acceleration in his suffering. If the pain and suffering he would have suffered anyway was made worse by the anguish of knowing that his disease could have been detected earlier, then he should be compensated for that. There is also the distinct possibility that the delay reduced his life expectancy in the following sense. It is possible that had he been treated when he should have been treated, his median life expectancy then would have been x years, whereas given the delay in treatment his median life expectancy from then is x minus y. This argument requires that the assessment of loss of life expectancy be based on median survival rates: i e those to be expected of half the relevant population at the particular time. If half the men with Mr Gregg's condition would have survived for x years or over with prompt treatment, and half would have survived for less than x years, then x is the median life expectancy of the group. If the same calculation of life expectancy from when he should have been treated is done in the light of the delay in treatment, the median life expectancy may have fallen. There might therefore be a modest claim in respect of the ‘lost years’.”

Dr Nathan, whom I found to be a most impressive witness, was asked to comment on this part of Professor James’ evidence. He agreed with the argument in principle, while adding (correctly, of course) the rider that he always tells his patients that each of them is an individual, not the median. He said that, just as he would expect the tumour to have thickened between March and October, so he would expect the degree of nodal spread to have increased. The patient’s prospects were worsened, not only in the sense of a reduced chance of survival beyond 10 years, but also in the sense of a reduced expectation of life. He drew my attention to the AJCC survival curve table for patients with stage III melanomas. This shows a median survival time from diagnosis and excision of a Stage IIIC melanoma of approximately 4 years, and a median of 7.5 years for a Stage IIIB melanoma.

Mr Kennedy submitted that it is too early to say what difference has been made in the Claimant’s case, and that this issue should be left to a further hearing. I indicated that I was not prepared to do that, since (a) a third hearing on liability in this litigation is unacceptable; (b) given the relatively modest damages recoverable by comparison with those at stake in the primary claim, it would be disproportionate; (c) in any event, it is unnecessary. The gap between the AJCC Stage IIIB and Stage IIIC curves continues at over 3 years for some distance on the chart. The Claimant has already done better than the median for a patient diagnosed at Stage IIIC, which suggests that he has a more combative immune system than most. There is every reason to suppose, as the AJCC tables show, that he would likewise have outperformed the median if he had been diagnosed at Stage IIIB.

I find that on the balance of probabilities the failure to diagnose the tumour in March 2006 has caused the Claimant’s life expectancy to be reduced by three years.

The Claimant is therefore entitled to judgment for damages to be assessed, but in respect of his alternative claim only. It is very much to be hoped that with the findings of fact made in this judgment and that of Mackay J the parties can now agree quantum. In the unhappy event that they cannot, quantum should be tried without delay. I invite submissions from counsel as to what directions would be appropriate.