PI Newsletter June 2019 – Delayed diagnosis in Cancer

By Andrew Arentsen

As medical science progresses, more sophisticated diagnostic tests reveal earlier stages of cancer and pre-cancerous cells. At the same time the efficiency of the treatment is improving. Advanced surgical techniques, precision radiotherapy and the increased personalizing of drug regimes have all increased the efficiency of cancer treatments. Immunotherapy (in which the treatment focuses on boosting and re-invigorating the immune system to fight the cancer) is becoming increasingly more common. The combined effect ultimately increases a patient’s chance of survival. 20 years ago, the chances of surviving after a diagnosis of cancer was about 20-30%. Now the prospects of survival are greater than 50%.

This places an increased strain upon primary NHS resources. More patients will now present with diagnosable early stage cancers. They may have modest symptoms. It is easy for a GP to diagnose a far more common, benign condition in the very short window that they have with the patient. A diagnosis can be missed and the opportunity for early treatment passes.

An average GP will only see about 6 patients a year who have cancer. They will often see 50-60 patients a day. Few symptoms are caused only by cancer. Almost all symptoms of cancer could be a symptom of cancer or could be (and are more likely to be) a symptom of something more benign. A GP is looking for a needle in a haystack. The 2015 NICE Guidelines suggest that the GP should make a specialist referral if there is more than a 3% chance that the patient’s symptoms could be caused by cancer.

Though referrals from primary to secondary care are up by 10% each year, the UK falls well behind other developed nations such as Sweden, Australia and Canada for cancer survival rates. GP’s are historically “Gate Keepers”. They are sifting patients to keep them out of hospital. There is also evidence that difficulties with access to diagnostic pathways mean that referrals are often avoided or are delayed.

With the huge increases in cancer caused by ageing and increased exposure to carcinogens, this is going to become a significant problem. The government has set a standard of 75% of cancer cases being diagnosed at stages 1 and 2. The general consensus is that early diagnosis saves lives. There is some interesting expert debate about this. Lead time bias is a major factor. If you diagnose cancer earlier then survival rate post diagnosis will inevitably increase even if treatment is ineffective. 20 years ago a man might be diagnosed with cancer aged 30. He then dies at 32. The survival rate post diagnosis is 2 years. Now he might be diagnosed with cancer when it is an earlier stage when he is 25. He might still die at 32. The survival rate has increased to 7 years post diagnosis but it has nothing to do with any treatment.

However broadly the experts cannot reconcile the importance placed on early diagnosis with the contention that early diagnosis and treatment makes no difference. They inevitably have to accept that as a general rule, early diagnosis and early treatment increases the rate of survival. A Claimant is on the front foot where there has been a delay in diagnosis which should not have occurred. The purpose of this article then is to consider the guiding case law and to consider the appropriate way to approach cases of this very particular type.
Gregg v Scott [2005] UK HL2

Gregg is the leading authority on cases involving missed diagnosis of cancer. It is the House of Lords’ decision that is definitive on the specific issues of causation. Mr Gregg developed a lump under his left arm as a consequence of non hodgkin’s lymphoma. In November 1994 he went to see his GP Dr Scott about the lump. Dr Scott concluded that the lump was a benign lipoma. The Trial Judge concluded that this was negligent because he was not entitled to exclude as a differential diagnosis the possibility of a more sinister pathology and he should have referred the Claimant to hospital for a specialist assessment.

In August 1995 Mr Gregg saw a new GP at another surgery. That GP concluded that the lump was probably a lipoma but he couldn’t be sure and referred Mr Gregg to the Lincoln County Hospital for further investigation. On 2nd November 1995 he underwent an urgent biopsy which confirmed that the lump was cancerous. In January 1996 Mr Gregg was admitted with acute and intense chest pain with the cancer having spread into the left pectoral region. He underwent chemotherapy and then radiotherapy. Although the tumour responded it did so incompletely. Mr Gregg then underwent high dose chemotherapy involving the elimination of certain types of cell, the harvesting of stem cells and the re-introduction of those cells into the body. He was discharged in September 1996 but in early 1998 he suffered a relapse when he developed a tumour in the right axilla. Statistically this gave him a very poor prognosis. Chemotherapy was commenced which was merely palliative. Thereafter there was no recurrence of the disease but the statistical probability of a fatal remission occurring was very high.

At first instance the Judge determined that the particular type of cancer suffered by Mr Gregg gave him a 42% chance of surviving for 10 years if it had been treated promptly upon the first diagnosis. The 10 year survival rate represents the best estimate of a full survival rate. Follow up evidence was not kept after 10 years. The chances of Mr Gregg now surviving 10 years having suffered 2 relapses were assessed at 25%. Thus as a result of the failure to identify the cancer at the correct time, Mr Gregg had suffered a very clear reduction in his prospects of survival from 42% down to 25%. However since the prospect of survival would always have been less than 50% this additional reduction in the chance of survival breath that prospect did not give rise to a claim. On the balance of probabilities Mr Gregg had not lost anything since he had nothing to lose.

The Court of Appeal and ultimately the House of Lords agreed with that assessment. There was no provision to claim a “lost chance” in clinical negligence claims. The Court distinguished other cases with a “loss of chance” such as Chaplin v Hicks [1911] 2 KB786 where Claimant was unable to attend a beauty contest. The Court could never know the outcome of the beauty contest had the Claimant appeared. It was therefore appropriate to treat her loss of chance as actionable damage.

Lord Hoffman denied that this approach was appropriate to the subject claim. The progress of Mr Gregg’s disease had a determinate cause. It was either inherent in his genetic make up or it was affected by subsequent events but medical science could not say that the outcome was random. The inability to establish that the delay in diagnosis caused a reduction in life expectancy cannot be remedied by treating the outcome as having somehow been indeterminate. The claim failed because the Appellant had not proved that it was more probable than not that, had there been no delay, he would not have suffered those consequences. The Appellant was not able to prove on the balance of probability that but for the delay in treatment he would have made a complete recovery

There were two powerful dissenting Judgments from Lord Nicholls and Lord Hope. This illustrated the concerns that troubled the Court. The Claimant had plainly lost something as a result of the breach of duty. His cancer had grown and the prospect of successfully treating the same had reduced. That might ultimately be the difference between his life and death. It seemed unfair to determine the case simply on the basis of whether his initial prospects of survival were statistically 51% or 49%. The dissenting Judgments each suggested that a significant reduction in the prospects of a successful outcome constituted a measurable loss for which the Appellant was entitled to be compensated. Neither Judge then considered how one would quantify that claim.

On the facts of Gregg, the court concluded that on the balance of probability the delay in commencing treatment had not affected the long-term course of the illness or the prognosis. However the delay may well have led to a more invasive treatment regime with additional pain, suffering and distress than had the diagnosis been made in time and the treatment been commenced promptly. Those additional consequences would have been recoverable but were not the subject of the appeal. Baroness Hale suggested that 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. It is possible that had he been treated when he should have been treated that his median life expectancy would have been x years whereas given the delay in treatment his median life expectancy is now x minus y. There might be a modest claim in respect of “the lost years”. None of these matters were explored before the Trial Judge since the focus was on whether the Claimant would have achieved a total cure.

The case of JD v Dr Melanie Mather [2012] EWHC 3063 did then involve an assessment of a further reduction from an already reduced life expectancy. It involved the failure to diagnose a malignant melanoma. The Claimant had become aware of the growth of a lesion in his groin in the middle of 2005. He attended his GP’s Surgery in March 2006 because the lesion “appeared to him to be growing, it was itching and it had bled when he scratched it.” The Defendant GP diagnosed the lesion as a seborrheic wart because it had the typical “stuck on appearance and demarcated borders.” She reassured the Claimant and he was discharged. The Claimant re-attended his GP on 16th October 2016 when he saw the Defendant’s Father. By that date the lesion was “huge” and the surface was removed by curettage. Subsequent histological examination demonstrated that the lesion was an ulcerated malignant melanoma. The Breslow thickness (the vertical distance from the surface of the skin to the lowest cancerous cell) was approximately 5 mm. Lymph node analysis identified metastatic melanoma.

In the light of earlier preliminary findings, the Defendant admitted breach of duty. She should have referred the Claimant for expert treatment on 8th March 2006 but she denied any causative damage flowing from that negligence. The trial in front of Bean J (as he then was) on the issue of causation.

Had the Claimant been referred to a specialist in March 2006 then he would have been seen within 2 weeks and the primary tumour would have been excised. If the tumour has not spread by that time then the surgical excision of the same is a complete cure. His case was that the excision and treatment would have carried a likelihood of care and ordinary survival. A supplementary report served just days before the trial, postulated an alternative case that the excision and treatment in March 2006 would have at least given the Claimant a longer period of disease-free survival. By trial the cancer was classed as Stage IV and had spread to the lungs. The prognosis was very poor indeed.

The case turned on the likely characteristics of the tumour in March 2006 compared with the characteristics that were actually found in October 2006 and the impact of any differences upon the ultimate prognosis. This depended upon the stage that the cancer had reached. The staging is set under the American Joint Committee on Cancer (AJCC) which used 4 predictive factors. The Breslow thickness of the primary tumour, whether the tumour was ulcerated, whether the cancer had reached 1 or more of the lymph nodes (and whether this was palpable or visible only under a microscope) and finally whether there has been any metastatic spread of the disease beyond the primary site.

Bean J determined that the Breslow thickness was between 3 – 4 mm in March 2006. On the balance of probability the tumour was already subject to malignant ulceration by that date. The lymph nodes were clinically abnormal and histologically positive from melanoma in October 2006 but of course there was no evidence as to their condition in March. He therefore needed to work back to determine the condition of the lymph nodes 7 months earlier. He accepted the Defendant’s expert evidence that it was rare for a tumour that has not already spread to give rise to palpable lymph node within 7 – 8 months and it is far more likely that the Claimant’s tumour had spread with microscopic regional lymph node involvement by March 2006. The result of his findings of fact were that the Claimant’s melanoma in March 2006 was Stage III B of the AJCC staging and by October 2006 it had reached Stage III C. The Claimant’s prospects of surviving a further 10 years were already less than 50% by March 2006 even with a diagnosis and a primary resection . He applied the decision in Gregg v Scott. The Claimant had not proved that the negligence had caused the failure to cure his cancer.

However the Claimant did succeed on his alternative case. The Claimant’s prospects had 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. The AJCC survival curve table for patients with Stage III melanomas showed a median survival time from diagnosis and excision of a Stage III C melanoma of approximately 4 years and a median survival time of 7.5 years for a Stage III B melanoma. Bean J found on the balance of probability that the failure to diagnose the tumour in March 2006 caused the Claimant’s life expectancy to be reduced by 3 years

Similar issues presented in the case of Loretta Oliver v Dr Gary Williams [2013] EWHC 600 QB. The Claimant attended her GP in December 2006 complaining of abdominal symptoms, bloating and diarrhoea and the Defendant made an urgent referral to the St Mary’s Hospital but the referral letter was lost at some point and no appointment was arranged. The Claimant’s case was that she was not told that she was being referred for urgent investigation and that had she been told she would have acted independently to check and secure an appointment at the St Mary’s Hospital.

The Claimant re-attended her GP in June 2007 and was again referred to the St Mary’s Hospital where a CT scan was performed on 5th July diagnosed ovarian cancer. On 19th July she underwent surgery. Widespread disseminated malignant deposits were identified which were between 0.5 – 1 cm in size. The Consultant Gynaecologist removed as much of the malignant deposits as possible and the Claimant underwent adjunct chemotherapy.

The Court found as a fact that the Defendant had not told the Claimant that he was urgently referring her for investigation at the St Mary’s Hospital. The Court concluded that had she been notified of that then she would have chased matters and obtained an appointment at the hospital or sought a private appointment and that the ovarian cancer diagnosis would have been made by late January/early February 2007. The breach of duty caused a delay in diagnosis and treatment of 5½ months.

The expert Gynaecologists agreed that a period of 5½ months would have made no difference to the named classification Stage of the disease or the treatment options. In February 2007 and in July 2007 the disease was at Stage III C. The management would have been identical. Surgery to remove as much cancerous tissue as possible followed by adjunct chemotherapy

The Claimant’s case was that the surgery in February would have led to less residual abnormal tissue remaining after surgery than was left following the surgery in July. This then translated into a diminution in her life expectancy.

The Court rejected the contention that the difference in surgery being undertaken in February compared with that undertaken in July translated into any reduction in life expectancy. There was no literature which satisfactorily established that leaving residual abnormal tissue where the maximum dimension of each tumour remained less than 2cm had any impact on life expectancy though the literature did support the contention that leaving aggregate material greater than 2cm was associated with a diminution in life expectancy. One might accept the reasoned explanation that the smaller the volume of disease that remained when chemotherapy begins the less likelihood of clone of resistant cells developing. This evidence was sufficient to conclude that on the balance of probability, in general terms, the less volume of cancerous tissue left after surgery the better the prognosis for survival. However there was no evidence as to what proportion of residual cancerous tissue would have been left in February as opposed to what was left in July and no evidence to suggest a reduction in life expectancy of between 12 – 18 months or at all. Whilst the volume of residual disease was a prognostic indicator, if one cannot assess the difference in volume of residual material there is no evidence to support the contention that the Claimant lost a chance of such an improvement. The Claimant needs to prove on a balance of probability that the delay of 5½ months made a material difference to her life expectancy, that was a measurable difference to her life expectancy. The case contrasted with that of JD (above) where the staging of the melanoma had changed and good quality statistical information was available to show the impact of a change in stages on medium life expectancy.

The Claimant was awarded a sum of £2,500 for a period of 5½ months of bloating and diarrhoea. She would not have suffered those symptoms had she undergone surgery and chemotherapy at an earlier stage. In addition the Claimant had suffered psychological injury. The psychological evidence was inadequate and did not distinguish between the psychological injury caused by the cancer diagnosis and treatment and the psychological injury caused by the delay in cancer diagnosis and the different treatment. The court concluded that the bulk of psychological symptomatology would have occurred irrespective of the delay in diagnosis however her anger, loss of control and belief that things may have been different but for the delay made a material contribution to the psychological injury. It would have awarded general damages of £15,000 for the psychological injury in full but awarded general damages of £5,000 for the psychological injury attributable to the breach of duty. It was a broad approach.