Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Agree with the idea - but disagree with the direction…
- Jennifer George (7 June 2002)
-
An example
- Dinkar D. Palande (8 June 2002)
-
More examples - with good reasons
- Leopold Reinecke (9 June 2002)
-
Commissioning organisations need estimates of cost effectiveness
- David W Black (11 June 2002)
-
Re: More examples - with good reasons
- Jennifer George (13 June 2002)
-
"Good intentions, double standards and bad science"
- Ross Camidge (29 June 2002)
-
Surrogate markers as clinical endpoints for cancer trials: currently a realistic option?
- Cornelis J A Punt (25 July 2002)
|
|
|||
|
Jennifer George, Associate Professor of Pharmacology St. John's Medical College, Bangalore - 560034, India.
Send response to journal:
|
Yes, endpoints should look at relief of symptoms & quality of life. But, equally, the relevance and significance of the results need to be clearly established. I find it difficult to agree (and so will most patients, I suspect), with the statement in the article: "Yet for an individual patient even a VERY short period without symptoms can be of value despite SERIOUS side effects"(emphasis mine). What needs to be remembered is that in actual practice, prediction of effect on a given patient is even more difficult than predicting effects on trial patients. Two important reasons are: 1) Trial patients are carefully chosen by strict adherence to well-defined inclusion and exclusion criteria. (Most cancer patients in the free-living population will probably be excluded from clinical trial studies - a major reason why recruitment takes ages). 2) While patients are in the trial, some degree of a "feel-good" factor operates as they are looked after extremely well in terms of psychological & emotional support from their doctors, free investigations, etc etc. (These 'goodies' are usually not available to cancer patients in the free-living population.) Hence, if a study result indicates only 25% or less improvement in symptoms/ progression, then that fact needs to be explained to the patient in very very clear terms. At the same time, the seriousness of the adverse effects and their consequences, also need to be just as clearly discussed. It is for the patient to make a decision on whether it is worthwhile to put himself through the therapy. On whom is the onus for such explanation? There lies the crux of the problem - doctors do not spend enough time reading up the evidence for themselves and then explaining it to the patient. If a doctor does that right, I'll wager hardly any patient will opt for a therapy which may provide a very short period of relief AND serious adverse effects. |
|||
|
|
|||
|
Dinkar D. Palande, Vicechair Lepra India 27,Advocate Chinnathambi Street Pondicherry India 605012
Send response to journal:
|
I agree fully. As Dr. Jennifer has said, taking the affected person in confidence and explaining the odds is essential. I have had similar experience with Tamoxifen for Cancer Breast,and had to smuggle it for the treatment of a close relative since the drug was not then in the market, pending approval. How to measure quality of life? apart from the recent developments, the grading by the affected person and her/his attendents subjectively is essential. |
|||
|
|
|||
|
Leopold Reinecke, Radiation oncologist Rand Clinic
Send response to journal:
|
Dr. Koopman's article is a timely one. (Refer to the news article in this issue of the BMJ on the drug now available for chronic myeloid leukaemia). Every oncologist in private clinical practice has from time to time to make a choice between a patient's possible survival and waiting for the outcome of clinical trials. When the patient is sitting in front of the doctor and there seems to be a logical reason to use a given chemical or biological agent which may give a good response - (good sound pathophysiological reasons) - it is difficult for the clinician to withold what is sound treatment based on first principles just because the "Phase II" study is the only one available; or the "Phase III" study has not yet accrued sufficient patients to draw any conclusive results. Provided the patient understands the reasoning of the oncologist - and that requires time and effort in an explanation; and the patient does not suffer from any harm from the intended "new" treatment. Also, in this sort of situation the disease is known to be uniformly fatal and the prognosis is zero. And if the Medical Aid Society accepts the rationale for the treatment that is to be given - Must the doctor and the patient then wait? I started using alfa interferon 5 to 10 years before there were adequate clinical data to substantiate its use in several diseases including myeloma, melanoma and non hodgkins lymphoma. Some of my patients are now 11+ year survivors; and one a 7 year survivor. Others are 5 year and more survivors...Eventually the clinical trials and data have caught up with sound basic medicine based on first principles. "Good treatment makes a patient better - bad treatment makes the patient sick" - a simple enough philosophy for general practice - and not completely out of place in oncology practice either. My patients would now have been dead and buried long ago before there was sufficient evidence for the use of this particular biological response modifier. A rethink about the criteria of sound physiological reasons for use in treatment is long overdue. As new agents become available now and in the next few years - this "paradigm" which is "established" (awaiting clinical trials) ...will come under increasing pressure from the patients and physicians alike. A valuable book illustrating the politics of "waiting" (1) is well worth reading. Reference: (1) Bazell, R. Her - 2 The Making of Herceptin a revolutionary treatment for breast cancer. Random House Inc. New York 1998:1-187. |
|||
|
|
|||
|
David W Black, Acting Director of Public Health Chesterfield PCT, Department of Public Health, Scarsdale, Newbold Road, Chesterfield, S41 7PF
Send response to journal:
|
Dear Sir Commissioning organisations, such as Primary Care Trusts (PCTs), wish to fund the introduction of cost-effective new treatments as soon as possible. This is consistent with their responsiblity to improve health, which they must do within a limited budget. Unfortunately, new drugs, including drugs for cancer are usually costly and licensed without adequate cost effectiveness data. Without reasonable estimates of cost-effectiveness it is inappropriate to divert resources away from interventions already known to be cost-effective. Sufficient research data to make estimates of cost-effectiveness are required before funding these drugs. |
|||
|
|
|||
|
Jennifer George, Associate Professor of Pharmacology St. John's Medical College, Bangalore, INDIA
Send response to journal:
|
Leopold Reinecke says, provided there are "good sound pathophysiological reasons [to use a new drug] AND the patient does not suffer from any harm from the intended "new" treatment", then doctors should go ahead & prescribe, without waiting for information from Phase III studies. I wonder which anti-cancer agent fulfils the second criterion - that of 'no harm at all'? Certainly, neither interferon alfa or herceptin fit the bill. New black box warnings continue to be released for both these agents! |
|||
|
|
|||
|
Ross Camidge, Clinical Lecturer in Cancer Therapeutics Edinburgh Cancer Centre, Western General Hospital, Edinburgh, EH4 2XU, UK
Send response to journal:
|
Few oncologists would disagree with Dr Koopmans1 desire to get ‘valuable and useful drugs licensed without undue delay.’ The long slow wait for drugs to show a survival benefit in phase III studies before they can be licensed feels even longer when your clinics are filled with needy patients brandishing promising phase II results downloaded from the internet. Yet although surrogates of benefit detectable at an earlier stage in oncology trials hold considerable promise for the future, none are well validated at present. So while Dr Koopmans accurately observes that a double standard is in operation when, for example, anti-obesity and anti-viral drugs are already being licensed on the back of studies using poorly validated surrogate endpoints, his solution to this inequality is lamentable. To commit the same scientific sin, by formally assessing anti-cancer drugs using potentially invalid data as a method of hastening the licensing process would simply allow doctors, health care providers and patients the dubious enjoyment of repenting the decisions at leisure. The organizational and financial backing of many large high-quality studies by the pharmaceutical industry is primarily to acquire product licenses. Gaining the same high quality data once a product has already been licensed, or even getting patients to enter into the appropriate control arms after a company has begun the widespread marketing of their product, may be impractical, occasionally even impossible. The difficulty in recruiting CML patients into any randomized study with a no imatinib (Glivec) treatment arm after the massive global publicity surrounding this product exemplifies this issue. Dr Koopmans’ other suggestion that side-effects are an inconvenient confounder of survival data, and therefore an irrelevance to the licensing procedure, also seems, at best, naive. There would be few benefits associated with licensing a new drug that admirably controls a surrogate endpoint of cancer, but the ‘confounding’ neutropaenia of the treatment kills 40% of the patients. 1. PP Koopmans. Clinical endpoints in trials of drugs for cancer: time for a rethink? BMJ 2002;324:1389-1391. |
|||
|
|
|||
|
Cornelis J A Punt, Professor, department of medical oncology University Medical Center St. Radboud, Dept. of Medical Oncology, PO Box 9101 6500 HB Nijmegen NL
Send response to journal:
|
Koopmans (1) holds a plea for the use of surrogate endpoints, such as biomarkers, response rate and quality of life data, for early licensing of anticancer drugs instead of the more time-consuming and more costly assessment of overall survival. He supports this by stating that cancer patients prefer symptom relief to prolongation of survival. This however does not apply to symptomatic cancer patients in general, and obviously not to the large number of symptom-free patients who want effective drugs to reduce their risk for recurrence (adjuvant setting). Yet there is general consensus that surrogate endpoints would greatly facillitate the selection of effective anticancer drugs. However no validated endpoints are currently available. Koopmans quotes response rate as a usefull surrogate endpoint for survival (2). However in the original publication the authors state that “in the context of individual trials, knowledge that a treatment has benefits on tumour response does not allow accurate prediction of the ultimate benefit on survival”. There are numerous examples of new drugs or schedules that produce high response rates in phase II trials but that fail the test in large phase III trials. A recent example in advanced colorectal cancer is the response rate of 76% with 5-fluorouracil and interferon-alpha in the initial trial (3), which created great optimism resulting in numerous phase II and III trials involving thousands of patients. However in the end no benefit in either response rate, toxicity, quality of life, or survival was observed (4). A valid surrogate marker could have prevented the exposure of these patients to a toxic, ineffective, and costly treatment, and this supports the plea of Koopmans. However, response rate clearly did not came up to this requirement. On the contrary, irinotecan has been approved as second line treatment for advanced colorectal cancer based on a significant survival benefit (5) but this drug would not have survived based on its response rate of 13%. Furthermore, for drugs that are less likely to induce tumor remissions such as cytostatic (as opposed to cytotoxic) drugs or immunomodulators such as vaccines, we are still stuck with time- consuming phase III trials as long as we have no validated biological markers. The pittfall of early licensing has been demonstrated by the monoclonal antibody edrecolomab which showed a significant improvement in overall survival versus observation when given as adjuvant treatment for stage III colorectal cancer, however in an underpowered study (6). It was claimed that the observed survival benefit was comparible to that of 5- fluorouracil /leucovorin which by then had become the standard, upon which this drug was licensed in Germany. However in a phase III trial with 2700 patients edrecolomab showed a significantly worse survival compared to 5- FU/Leucovorin (7). In the meantime many patients will have received this drug and were thus exposed to an inferior treatment. This would have been prevented if the licensing had been based on a more rigorous test. Lastly, if survival data should be collected in the postmarketing period, as Koopmans suggests, who should pay for these trials? Clearly the pharmaceutical industry will have very limited interests in these expensive studies when their drug is on the market and eligible for reimbursement. Moreover, aggresive marketing strategies may be more effective than scientific data from clinical studies, such as the case for adjuvant treatment with high-dose interferon-alpha for melanoma which is still wrongly being considered by many oncologists as a drug that prolongs survival (8). Those oncologists that do know often have great difficulties to explain their patients why they do not prescribe this drug despite its registration and its promotion on the internet. The search for surrogate endpoints should definitely be continued, but they currently cannot yet be relied upon in the introduction of new anticancer drugs to the market. References 1. Koopmans PP. Clinical endpoints in trials of drugs for cancer: time for a rethink? BMJ 2002;324:1389-91. 2. Buyse M, Thirion P, Carlson RW, et al. Relation between tumour response to first-line chemotherapy and survival in advanced colorectal cancer: a meta-analysis. Lancet 2000;356:373-8. 3. Wadler S, Schwartz EL, Goldman M, et al. Fluorouracil and recombinant alfa-2a-interferon: an active regimen against advanced colorectal carcinoma. J Clin Oncol 1989;7:1769-75. 4. Thirion P, Piedbois P, Buyse M, et al. Alpha-interferon does not increase the efficacy of 5-fluorouracil in advanced colorectal cancer. Br J Cancer 2001;84:611-20. 5. Cunningham D, Pyrhönen S, James RD, et al. Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 1998;352:1413-8. 6. Riethmüller G, Schneider-Gädicke E, Schlimok G, et al. Randomised trial of monoclonal antibody for adjuvant therapy of resected Dukes’ C colorectal canrcinoma. Lancet 1994;343:1177-83. 7. Punt CJA, Nagy N, Douillard JY, et al. Edrecolomab (17-1A antibody) alone or in combination with 5-fluorouracil and leucovorin in the adjuvant treatment of stage III colon cancer: results of a multinational phase III study. Lancet 2002; in press. 8. Punt CJA, Eggermont AMM. Adjuvant interferon-alpha for melanoma revisited: news from old and new studies. Ann Oncol 2001;12:1663-6. Prof. dr. C.J.A. Punt University Medical Center St. Radboud Deparment of Medical Oncology P.O. Box 9101 6500 HB Nijmegen The Netherlands c.punt@onco.azn.nl |
|||