Rapid Responses to:

ANALYSIS: Tom Treasure, Martin Utley, and Ian Hunt When professional opinion is not enough BMJ 2007; 334: 831-832 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

When professional opinion is not enough

Alan Leaman   (24 April 2007)

Resection of Pulmonary Metastasis in Colorectal Cancer; a known unknown clinical decision for the oncologist.

SWETHAJIT BISWAS   (24 April 2007)

Professional opinion is not enough

Penella J Woll, Jo Hornbuckle, Sarah J Danson, Patricia M Fisher, Bernadette Foran, Matthew Q Hatton   (27 April 2007)

When professional opinion is not enough

Graeme J Poston, O James Garden, John N Primrose, Myrrdyn Rees   (1 May 2007)

Authors' reply

Tom Treasure, Martin Utley   (8 June 2007)

When professional opinion is not enough 24 April 2007
Alan Leaman, Consultant in Emergency Medicine Princess Royal Hospital, Telford Send response to journal: Re: When professional opinion is not enough	Professor Treasure and his colleagues have provided another example of flawed guidance from NICE (1), but the fundamental question here is whether this organisation should be allowed to dictate medical practice in the UK. The Department of Health is now assessing hospitals on the extent to which they comply with NICE guidance, yet much of this guidance is controversial and has a weak or non existent evidence base. Furthermore because NICE is a government body, doctors who choose to ignore it’s advice may find themselves in un-chartered legal territory (2). NICE should be seen for what it is - another arm of an over- controlling government. NICE does not have a monopoly of wisdom, and the interests of patients are best served if doctors question it’s guidance, and continue to make their own decisions about patient care. AM Leaman Consultant in Emergency Medicine (1)Treasure T, Utley M, Hunt I. When professional opinion is not enough BMJ 2007; 334: 831-832 (2)Baines P. NICE head injury guidelines; review of the legal mandate Emerg Med J 2005; 22:706-709 Competing interests: None declared
Resection of Pulmonary Metastasis in Colorectal Cancer; a known unknown clinical decision for the oncologist. 24 April 2007
SWETHAJIT BISWAS, Specialist Registrar in Medical Oncology/CR UK Clinical Research Fellow John Radcliffe Hospital, Oxford, OX3 9DU Send response to journal: Re: Resection of Pulmonary Metastasis in Colorectal Cancer; a known unknown clinical decision for the oncologist.	Professor Treasure's recent commentary on the role of pulmanoary metastatic resection in colorectal cancer highlights the lack of randomized control trials in surgery. This is however not an isolated surgical problem. In the era of multi-discplinary care in cancer medicine, this is a pressing issue for oncologists as well. Most cancer physicians who refer such patients for pulmonary resection to thoracic surgeons are acutely aware that these patients represent a unique subgroup of patients with metastatic disease and may indeed be self-selecting due to the "good biology" of their respective tumours. Indeed, this is where the problem lies. Most of our clinical judgements in referral to the thoracic surgeon is based on our own ancedotal past experiences in this setting and the current morbidity of the patient when they attend. There is no substantial evidence base for what we do in terms of the oncologists referral to the thoracic surgeon and subsequently what the thoracic surgeon does to the patient. All of this may feel, that we, as the medico-surgical team, are doing something positive for the patient and indeed the patient feels that removing the isolated metastatic lesion can only be a good thing. The reality is that we just don't know or as Donald Rumsfeld would put it, this particular course of patient management is "a known unknown". Do we watch and wait (an un-nerving experience for the patient), intervene when pulmonary symptoms develop or remove the lesion when first CT imaged/ biopsied? My gut instinct (therefore non-evidence based) is that these patients with isolated pulmonary metastasis may have a genetic/proteonomic signature that not only favours pulmonary tropism for metastasis engraftment and development, but this is enextricably linked to overall survival. Given the limited numbers of patients who develop surgically amenable isolated metastatic lesions of this type, only a prospective trial involving DNA microarray and serum protenomics from patients with primary colorectal cancer and subsequent similar analysis on their resected metastatic deposits could only tease out the dominant relevant molecular pathways in this sub- group of patients and this will also elucidate whether advances in adjuvant or first line metastatic chemotherapy regimes, in colorectal cancer over the last several years have altered the biology of the disease in a small group of patients, such that smaller, more operable lung lesions are developing. If this is the case, we need to know what chemotherapy combination therapies these are and can we predict at the outset of the disease, the development of future lung metastasis, as there maybe a perculiar lung metastatic genetic signature within the primary colorectal tumour. Competing interests: None declared
Professional opinion is not enough 27 April 2007
Penella J Woll, Professor of Medical Oncology University of Sheffield & Weston Park Hospital, Sheffield S10 2SJ, Jo Hornbuckle, Sarah J Danson, Patricia M Fisher, Bernadette Foran, Matthew Q Hatton Send response to journal: Re: Professional opinion is not enough	We fully support Treasure and colleagues (1) in resisting the acceptance of unproven surgical (and medical) interventions. Pressure to resect lung metastases from colorectal cancer is an increasing problem as international guidelines for surveillance of colorectal cancer patients now recommend CT scan of the chest and abdomen annually for 3 years in “high risk” patients eligible for curative intent surgery (liver or lung resection) (2). This increased surveillance of the chest increases the detection of early lung metastases. The rationale for this is weak – liver resection may have a role in the management of regional tumour spread but lung resection is being used for systemic tumour spread. In other tumour types (eg. breast cancer, melanoma), there is good evidence for resecting regional spread, but systemic treatments are required for systemic disease. There is no evidence that colorectal cancer is biologically different from other cancers. The most important differential diagnosis for patients presenting with a single pulmonary nodule is of a primary lung cancer. Treasure et al point out that few of the case series for metastectomy in colorectal cancer include the denominator – the total number of patients operated. Importantly, none states what proportion turned out to be primary lung cancers. Patients with a single pulmonary nodule should be assessed and treated for a possible lung cancer. All others should be considered likely metastases and advised that surgery is of no proven benefit. It is interesting to contrast the high level of evidence required for the introduction of medical treatments for metastatic cancer, where at least one randomised controlled trial showing a survival advantage is needed to obtain a licence for a new drug, and several to obtain NICE approval. It seems that surgical treatments can be introduced without any formal evaluation. We fully support these authors in calling for the development of a formal evaluation of pulmonary metastectomy in a clinical trial. References (1). Treasure T, Uttley M, Hunt I. When professional opinion is not enough. Br Med J 2007; 334: 831-2. (2). Desch CE, Benson AB, Somerfeld MR, Flynn PJ, Krause C, Loprinzi CL, et al. Colorectal cancer surveillance: 2005 update of an American Society of Clinical Oncology Practice Guideline. J Clin Oncol 2005; 23: 8512- 8519. Competing interests: None declared
When professional opinion is not enough 1 May 2007
Graeme J Poston, Consultant Hepatobiliary Surgeon University Hosppital Aintree, Liverpool L9 7AL, O James Garden, John N Primrose, Myrrdyn Rees Send response to journal: Re: When professional opinion is not enough	Dear Sir, When professional opinion is not enough (BMJ 2007; 334: 831-2) As the authors of the systematic review of evidence for surgical resection of colorectal liver metastases1 quoted by Treasure and colleagues2, and also of the UK Guidelines for resection of colorectal liver metastases3, we wish tocontribute to the debate on surgical resection for colorectal pulmonary metastases. We have known the natural history of under-treated metastatic colorectal cancer for over a decade4. Very few (<1%) will still be alive 5 years after diagnosis. Against this is the 5-year survival of 30- 50% reported in the surgical literature following hepatectomy for liver metastases5. Treasure and colleagues make salient points regarding the selection of patients with ‘good’ tumour biology, and the impact of this biology on the natural history of the disease. Similarly their argument about the evidence base for lung resection is cogent. Indeed, only 45 publications are identified in PubMed under ‘resection of colorectal pulmonary metastases’ and of these only 11 would meet the criteria of our own systematic review (>100 patients, >1 year follow up, prospectively collected survival data). We look forward to the publication of the authors’ review, but a quick overview of these papers suggests a mean 5- year survival following lung resection of 30-40%, although interestingly one of the larger series (239 patients from 2 centres) reported a 5-year survival of only 11%6. We would caution their enthusiasm for the interesting proposal for a clinical trial to try and resolve the benefits of pulmonary resection. As the authors rightly point out, an understanding of tumour biology is paramount to this question. However, the argument that the liver is the first ‘filter’ of metastatic disease in colorectal cancer is naive. Colorectal cancer that is metastasising via lymphatics (Stage 3 primary) will travel via the pre-aortic nodes to the thoracic duct to reach the systemic circulation and the lungs, so by-passing the liver. Similarly, rectal cancer invading the pre-sacral venous plexus will embolise from the internal iliac veins to the systemic circulation. These methods of spread may explain the observation in 1578 patients who had undergone resection of primary colorectal cancer, in one single centre, that 117 (11%) of 1013 rectal cancer patients developed pulmonary metastases, compared to only 20 (3%) of 565 patients with colonic primary tumours7. Clearly, any proposed study will need to stratify very carefully for many such confounding variables. The second major issue is the impact of contemporary chemotherapy (and possibly biotherapy). Most trials currently reporting on the treatment of non-operable disease show median survivals after commencement of first-line treatment exceeding two years. However, this might be overcome if the primary end-point for the proposed study was 3-year progression free survival. Lastly, the authors conclude that the trial is feasible and recruitment possible since, ‘when there is uncertainty, unbiased allocation is justified and is as rational as either alternative. The informed public understands that’. The EORTC 40004 study attempted to answer a similarly vexing question in the management of advanced colorectal cancer (the benefit of radiofrequency ablation [RFA] of unresectable colorectal liver metastases) and the trial design was almost identical to that proposed by the authors: randomisation to chemotherapy alone versus chemotherapy plus RFA. The trial was launched 4 years ago across Europe as a randomised Phase III with the objective of recruiting 450 patients (225 in each arm) and with multi-institutional backing (including CRUK). The trial terminated two years later with only 75 patients recruited and funding support was withdrawn.. The problem was that simply, ‘the informed public’ couldn’t understand the randomisation and the uncertainty. The solution to the question posed by Treasure and colleagues would probably be better achieved by a prospective pan-European registry of patients undergoing resection of colorectal pulmonary metastases. Graeme J Poston University Hospital Aintree, Liverpool O James Garden Regius Professor of Surgery, University of Edinburgh John N Primrose Professor of Surgery, University of Southampton Myrrdyn Rees Royal North Hampshire Hospital, Basingstoke 1. Simmonds P, Colquitt J, Primrose JN et al. Surgical resection of hepatic metastases from colorectal cancer: a systematic review of published series. Br J Cancer 2006; 94: 982-99 2. Treasure T, Utley M, Hunt I. When professional opinion is not enough. Brit Med J 2007; 334: 831-2 3. Garden OJ, Rees M, Poston GJ et al Guidelines for resection of colorectal cancer liver metastases. Gut 2006; 55: Supplement 3. 4. Rougier P, Milan C, Lazorthes F et al. Prospective study of prognostic factors in patients with unresected hepatic metastases from colorectal cancer. Brit J Surg 1995; 82: 1397-400 5. Adam R, Aloia T, Figueras J et al. Livermetsurvey: analysis of prognostic clinico-pathologic factors in 3216 patients with colorectal liver metastases. HPB 2006; 8(2):52-3 6. Regnard JF, Grunenwald D, Spaggiari L et al. Surgical treatment of hepatic and pulmonary metastases from colorectal cancer. AnnThorac Surg 1998; 66: 214-8 7. Pihl E, Hughes ES, McDermott et al. Lung recurrence after curative surgery for colorectal cancer. Dis Colon Rectum 1987; 30: 417-9 Competing interests: None declared
Authors' reply 8 June 2007
Tom Treasure, Professor of Cardiothoracic Surgery Guy's Hospital,London SE1 9RT, Martin Utley Send response to journal: Re: Authors' reply	We welcome the responses to our work and the opportunity to comment. Before discussing the points that relate specifically to our paper, we entirely concur with Professor Woll and colleagues concerning management of an isolated and undiagnosed pulmonary nodule. The starting point of our necessarily short piece is a premise of proven colorectal metastasis. We also accept that, seemingly, more formal evidence is required for the introduction of pharmacological compared to surgical interventions. The question we pose in our paper is what evidence is required for a practice to persist. In commenting that a statement in NICE guidelines rests on a single not directly relevant paper we were not intending to fuel the debate of whether it is reasonable for there to be an organisation that collates, synthesises and disseminates evidence related to the effectiveness of medical or surgical interventions. Our concern is the paucity of evidence available for an operation as widely practiced as lung metastasectomy. Case series are the commonest form of evidence for cancer surgery and we have recently published a paper in which we discuss a number of pitfalls related to their interpretation. [1] We drew examples from reports of surgery for mesothelioma but could equally well have illustrated our points with examples from reports of other types of surgery. Graeme Poston and colleagues state that in 1996 the five year survival amongst patients that were “under-treated” (no prior value judgements there then) for colorectal metastases was considerably less than the survival in 2006 amongst patients that had liver metastases resected. We respect their caution in not claiming that this difference in survival is due to the surgery rather than clinical selection, self- selection, or lead-time shift. In passing, the argument of the liver being a first filter may be naïve but we are keen to point out that it is not our argument; we quoted it as one of the arguments made in support of resecting liver metastases. In the case of lung metastases, we do not doubt that patients that have metastases resected survive longer than an unselected group of patients with advanced colorectal cancer. The question is whether this survival difference is conferred by the act of surgery or is associated with the fact of surgery (i.e. a clinical team have deemed the patient a likely survivor). Registry data alone cannot be used to answer this question, nor whether any genuine survival difference due to surgery is worth the attendant morbidity. In his exposition of the genuine dilemma presented by patients with lung metastases, Dr Biswas suggests that, at times, major surgery may be preferred to inaction. Patients may assume that removal of a metastasis “can only be a good thing” but it is the duty of health professionals to ensure that the patients know that this is not necessarily the case. There are inescapable harms. Justifying an operation in this context for psychological reasons is something we would question [1]. [1] Treasure, T and Utley, M. Ten traps for the unwary in surgical series: A case study in mesothelioma reports. J Thoracic and Cardiovascular Surgery 2007;133:1414-8. Competing interests: None declared