Japanese Journal of Clinical Oncology

Department of Surgery, Leiden University Medical Center, Leiden, The Netherlands

INTRODUCTION

The first randomized clinical trial was carried out fifty years ago, when the efficacy of streptomycin in the treatment of tuberculosis was demonstrated (1). Since then, many, mainly medical, randomized trials have been conducted. The contribution of surgical randomized trials has been relatively low, although results from randomized trials are the only results used for `evidence based medicine'. In 1996, Richard Horton, editor of the Lancet, wrote a commentary entitled `surgical research or comic opera: questions, but few answers' (2). He reported that only 7% of papers published in nine general surgery journals were randomized trials and almost half of the publications were case studies and he therefore concluded that surgeons did not seem to see research as a key issue in their practice. Obviously he received a lot of criticism from the surgical community (3). Are surgical trials impossible to conduct or are surgeons or patients unwilling to participate?

STANDARDIZATION

Randomization of patients to evaluate surgical procedures involves problems in addition to those associated with medical experimentation. Surgery, unlike a pill, is not a standardized, reproducible entity, but rather a unique product whose details are defined by variables which include, for example, the skill of the surgeon. The skill level will not only vary among surgeons, but will increase for the same surgeon as he gains experience (4). Furthermore, surgeons with specific interest will perform better (5,6). These surgeons are also well-disposed to develop new techniques in their own center and subsequently analyze their series. This is one of the reasons why so many informative non-randomized hospital or personal series are published. New drugs on the other hand are developed by pharmaceutical companies and therefore from the start tested in the proper way, through phase I, II and III trials.

It is difficult to conduct a randomized trial in an evolutionary phase of a new operation (7,8). A prerequisite for a trial is that the participating surgeons are equally conversant with both techniques. An alternative to this problem could be the use of a randomized-surgeon design where groups of surgeons will perform conventional surgery and other groups of surgeons will perform the new type of surgery (8,9). However, a main objective of performing surgical trials is the improvement of outcome, which would occur at a slower pace if surgeons keep using conventional techniques.

Surgeons, training and infrastructure to perform surgical trial treatments need to be defined in detail in a protocol. Not only the type of surgery, but also the type of reconstruction, the treatment of complications and the grading of complications. A new surgical technique should be introduced at several levels, e.g. by booklets, videotapes, workshops and at the dissection table by instructor surgeons.

QUALITY CONTROL

In surgical trials, the level of training and expertise of those performing the procedures must be comparable and should always be documented (9). Documentation should be done in the standard surgical report as well as in detailed case report forms. After introduction of the new technique, operations of patients in trials should be attended by an instructor surgeon, not only to teach the local surgeon, but also to guarantee the standardization and quality of the operation (10). Quality control involving surgical treatment in multi-institutional cancer trials is also important since the results of adjuvant therapy might be obscured or overestimated by inadequate surgery or pathologic examination of the specimen. In rectal cancer, 27% local recurrence was observed in a surgery alone group versus 11% in combination with preoperative radiotherapy (11). In stage II and III breast cancer, mastectomy plus CMF gave a local recurrence rate of 32% versus 9% in the mastectomy plus CMF plus postoperative irradiation group. Both studies showed that the radiotherapy group also had a better chance of survival. In those two studies published in the New England Journal of Medicine (11,12) the surgery without radiotherapy group has recurrence rates which for most well trained surgeons would be unacceptable. Most trials, especially those evaluating systemic treatment do not use clear definitions or descriptions of the procedures. McArdle and Hole posed in 1991 `Some surgeons perform less than optimal surgery; some are less competent technically than their colleagues; and some fail to supervise surgeons in training adequately0If by more meticulous attention to detail the results of surgery could be improved, and our results suggest that this would not be difficult, the impact on survival might be greater than that of any of the adjuvant therapies currently under study' (5).

In The Netherlands, we are just conducting a rectal cancer trial where we question the use of preoperative radiotherapy in combination with Total Mesorectal Excision (13). A pure surgical study of conventional surgery versus TME surgery was not feasible: many hospitals were already engaged to the use of preoperative radiotherapy and would therefore not participate, what is `conventional' surgery, the problem of the randomized-surgeon design and most importantly the non-randomized evidence of the superiority of TME surgery over conventional surgery. If our study shows no difference in local recurrence rates, we can conclude that preoperative radiotherapy does not add to TME surgery. Therefore, costs will be lower, quality of life will be better and perhaps postoperative complications will be less. If preoperative radiotherapy reduces local recurrence rates, cost-effectiveness and quality adjusted life years will then become important factors for the yes or no standard use of preoperative radiotherapy.

TRIAL PARTICIPATION

Trial participation is dependent on physician and patient participation. In an emotional period after diagnosis of cancer where lots of information have to be absorbed, the doctor informs the patient on treatment and possible trial participation. In a busy practice, physicians have little time to ask for written informed consent and this might be a threshold to ask patients, especially when the trial involves acute patients. A written patient information sheet is important, but detailed personal, oral information should always prevail. This information can then be adapted to the patients level of understanding.

Furthermore, all the paperwork does not motivate to accrue more patients. Datamanagers for administrative and organizational support and research-nurses for the arrangement of informed consent might unburden the surgeon (14). A win-win situation in which the physician receives reports on their personal series might also stimulate them to fill in the forms and might also stimulate participation. Motivation is important and low accrual is more due to the doctor than to the patient. The surgeon should get the feeling of `missing the boat' if he does not participate. It is even suggested that survival of participants in clinical trials in oncology is higher than of those treated outside a trial structure (15).

In The Netherlands, an intensive consultant system is organized through comprehensive cancer centers. Physicians from university hospitals visit the community hospitals at their oncology meetings and stress the importance of trial participation, especially since the majority of primary operable patients are treated in those community hospitals. Yearly, approximately 1750 new rectal cancer patients are operated on with curative intent. In the second and third years of the trial, approximately 26% of all patients were randomized into the trial, without taking into account the exclusion criteria and non-participating hospitals. Our trial has no upper age limit, since otherwise with the aging population increasing numbers of patients could not participate and trial results would be less generalizable. The performance status of each individual patient is more important. In the future, more and more patients or relatives will browse the Internet, seeking for information about their disease and possible treatments. Patients might even ask their physicians about ongoing trials.

New drug trials are extensively sponsored by pharmaceutical companies. Surgical trials, however, have enormous difficulties in obtaining funding. In The Netherlands, we are relatively fortunate to have two major grant-giving bodies for clinical trials: the Dutch Cancer Foundation, which only supports on-site and central datamanagement and the Dutch Health Council, where also financial support for salaries of co-ordinators and e.g. reimbursement of expenses for pathologists and instructor surgeons can be applied for. Both organizations will only support randomized trials. Prospective registration studies are therefore hard to run.

CONCLUSIONS

Results from randomized trials provide the best scientific evidence of therapy efficacy or inefficacy. Also in a busy surgical practice, it is possible to conduct or participate in those trials. Practical information on the conduct of two successful surgical trials in the Netherlands will be given elsewhere in the journal. A pro-trial attitude in patients and doctors should be encouraged. Therefore, a close contact between an active co-ordinating datacenter, including co-ordinators for separate disciplines, and all participating physicians is essential for the well conduct of a quality controlled multicenter, multidisciplinary trial. Continuous enthusiasm can be kept by the organization of regular workshops, distribution of newsletters and trial up-dates at scientific meetings. The best way by which trial results are really translated to the surgical community as a whole are by conclusive trials with a fast accrual over a limited period of time. Therefore, medical and lay community mentality should better turn to why not exclude only when inclusion is excluded?