Nordic Myeloma Study Group, the first 15 years: scientific collaboration and improvement of patient care
Jan Westin, Stem Cell Center, BMC (B 10), University of Lund, SE-221 84 Lund, Sweden
Tel: + 46 46 222 75 74
Fax: + 46 46 222 36 00
Abstract: The accomplishments of the Nordic Myeloma Study Group (NMSG) during its first 15 yr are briefly surveyed, together with a discussion of principles guiding the group's clinical trials and of problems that need to be addressed in coming years. The group has so far carried out 12 clinical trials, comprising more than 2500 patients, spanning from minor phase II to large randomised phase III trials. At the time of writing, two randomised trials are running (comparing two doses of i.v. pamidronate, and melphalan-prednisone (MP) vs. MP-thalidomide to elderly patients). The group has strived for a simple organisation with much responsibility delegated to regional coordinators (Denmark 3, Norway 5, Sweden 5). With regard to trial design, the group has considered it important that studies are based on sound scientific questions, are simple to handle for the participants, population based, investigator initiated, include quality of life and health resources assessment as end-points, and can be used as basis for diverse scientific spin-off projects. Like other clinical trial groups, NMSG faces a number of challenges in coming years. The financial situation for independent investigator-initiated trials is far from satisfactory, especially with regard to the resource-consuming implementation of more stringent good clinical practice rules and ethical committee demands. NMSG has also encountered increasing difficulties in recruiting patients to recent trials, partly because of problems related to participating physicians (lack of support, laborious paper work, insufficient credit for participation). Solutions to these problems have to be found if industry-independent clinical trial groups are to survive.
The last decades have seen an increasing interest from clinicians and clinical scientists in the field of haematology to join forces in collaborative groups, focusing on disease entities or on diagnostic or therapeutic principles. Many such groups have contributed substantially to our knowledge of haematological disorders, including multiple myeloma, by introducing new therapies, establishing diagnostic and prognostic classifications and in the standardisation and quality assurance of different aspects of patient care.
The Nordic Myeloma Study Group (NMSG) was founded in late 1987, and has thus existed for a little more than 15 yr. Being instrumental in its birth and further development we have seen it grow from the first early steps of tentative contacts between just a handful of people into a mature collaborative clinical and scientific network, comprising today almost all clinics treating myeloma patients in Denmark, Norway and Sweden. We have found it pertinent at this moment to try to summarise our experiences of the build-up of the group and to review what with time has been established as the policy of the group regarding clinical trials, patient management and scientific spin-off project issues. It is our definite impression that establishment of the NMSG – as one of the first Nordic cooperative efforts within haematology – and the results achieved by the group have had a substantial positive effect not only on the quality of care of multiple myeloma patients in our region, but also on the interest for this disorder among Nordic clinicians and collaborating scientists.
How it started
In the middle 1980s much interest was focused on attempts to improve the chemotherapy in multiple myeloma by applying similar principles, that seemed successful in acute leukaemia and malignant lymphoma. This development was received with great hope and enthusiasm, there was a renewed interest in the disease, and almost simultaneously local initiatives, focusing on plasma cell disorder epidemiology and treatment, were started in all our three countries. In Copenhagen a small phase II study was initiated 1984 to evaluate if mitoxantrone (thought to be less cardiotoxic) might be substituted for doxorubicin in a vincristine-doxorubicin-dexamethasone-like regimen (mitoxantrone-vincristine-prednisolone, NOP), given to relapsing and refractory patients (1); in southern Norway (Helseregion I) a population-based study of myeloma incidence was started in 1985 (2); and in western Sweden Regional Guidelines for diagnosis and therapy of multiple myeloma had been established in 1986, and formed a natural base for two phase III trials (3, 4).
Presentations of data from these studies led to a close interaction and exchange of experience between the involved groups, and within a short time to the foundation of a formal (but still very informal) study group (NMSG). The main incentive for us who started the group was the necessity of a broader base for performing clinical trials, but we also felt the need to create a Scandinavian or Nordic forum, where we could meet and discuss common myeloma problems.
The primary task for the group became to perform a larger phase II study of the NOP regimen, and this first NMSG protocol (#1/87) (Table 1) succeeded to recruit 92 patients from 23 centres in Denmark, Norway and Sweden (5). In a follow-up study of this drug combination (#2/87) mitoxantrone was administered as bolus injections (6), and later a randomised phase III study of NOP vs. melphalan-prednisone (MP) as induction therapy (#3/89) was performed in elderly patients (7). Although the NOP regimen can be used as an alternative relapse therapy it was shown not to be better than MP as induction therapy. Performing these three studies gave us the experience of running multicentre, multinational protocols and welded the participating centres firmly together. We found it easy and rewarding to work with each other, the atmosphere of the group and of its meetings was informal and devoid of prestige, and we were all very satisfied that we had shown Nordic cooperation to be possible in the haematological area.
Table 1. Nordic Myeloma Study Group (NMSG) protocols 1987–2004
| 1||1987||II||NOP to refractory patients||92||23||Gimsing et al. (5)||NOP effective and tolerable as alternative relapse therapy|
| 2||1988||II||NOP-bolus to refractory patients||58||21||Wislöff et al. (6)||Mitoxantrone and vincristine in NOP regimen can be given as bolus injections|
| 3||1989||III||NOP vs. MP as induction therapy to elderly patients||151||16||Keldsen et al. (7)||NOP more toxic and not more effective than MP|
| 4||1990||III||MP vs. MP-interferon as induction and maintenance||583||107||Hjorth et al. (13)||Addition of IFN to MP during induction, maintenance and responding relapse prolongs time to progression by about 1.5 yr but not overall survival|
| 5||1994||III:H||High-dose melphalan with stem cell support <60 yr||274 (vs. H 274)||14||Lenhoff et al. (14)||HD melphalan + ASCT significantly improves overall survival compared with highly representative historical controls|
| 6||1996||II||Idarubicin as single drug to relapsing/refractory patients||30||11||Rödjer et al. (41)||Single drug idarubicin to relapsing or refractory patients has negligible effect|
| 7||1998||III:H||High-dose melphalan with stem cell support >65 yr||538 (vs. H 243)||14||Lenhoff et al. (42)||HD melphalan + ASCT may be offered to selected patients 60–65 yr of age with results comparable with patients <60|
| 8||2000||III||Pamidronate 30 vs. 90 mg||Goal = 500||50||–|| |
| 9||1999||–||Clinical validity of stem cell quality programme|| || ||Johnsen H (in prep.)|| |
|10||1999||II||Thalidomide to relapsed/ refractory patients||65||30||Waage et al. (43)||Thalidomide effective as salvage therapy. Positive effect can be seen early|
|11||2001||III||VAD vs. CyDex as induction before high-dose therapy||315||15||Mellquist UH, Lenhoff S, Hjorth M, Holmberg E, Tangen JM, Johnsen H (in prep.)|| |
|12||2002||III||MP vs. MP-thalidomide as induction to elderly patients||Goal = 800||40||–|| |
The group's fourth protocol (#4/90) was the hitherto most ambitious one, comparing in a large randomised phase III study traditional MP vs. MP plus alfa-interferon (IFN), given from the time of diagnosis through responding relapse. Against the background of promising early results of IFN therapy in myeloma (8–10) the study was met with great interest. It was easy to get participation and recruit patients from almost all hospitals in Denmark, Norway and Sweden, together with two centres in Swedish-speaking Finland and one in Iceland (finally n = 107). However, some invited centres chose to stay independent and continue their local activities.
The central organisation was initially very simple with a chairman of the running protocol, also being chairman of the NMSG, a secretary, a national coordinator for each country, and a number of regional coordinators (Denmark 3, Norway 5, Sweden 5, Finland 1). Together they formed the Steering Group, meeting twice a year, where all important decisions regarding the group's trials and other activities were made. The regional coordinators, each responsible for his/her university hospital region, were intentionally given a key role; they had to represent and promote the study, to monitor the performance of the local hospitals, to arrange regional workshops, where the progress of the study could be presented and discussed, and to be alert to problems.
In later years, when the activities of the group have expanded, the organisational structures have had to be more elaborate: by-laws of the group have been formulated and formally approved, an executive committee of eight persons formed, chairpersons designated for each running protocol, and several standing or ad hoc subgroups elected for special tasks, e.g. preparing new protocols and guidelines, spin-off projects, quality control. The structure of the group has hereby become more complex, and sometimes more difficult to survey, but a firm and open-minded leadership has so far spared us serious conflicts and prestigious power struggles. The chairmanship has been rotated every 3 yr among the participating countries. The fact that we have been able to conduct all discussions and use working material in our native languages has contributed to the feeling of togetherness within the group. An ambition has also been to increase the sense of participation by distributing duties and responsibilities among the steering group members.
The group has until now (September 2004) completed eight studies. The results from six of them have been published; from two recent trials results are under preparation for publication (Table 1). Two protocols are running and a third will start in October 2004. The trial design spans from a small phase II study (of single drug idarubicin), to several large randomised phase III trials, and two controlled studies of high-dose melphalan with stem cell support, using patients from our own previous studies as historical controls.
Development of a new protocol
The procedure, by which a new NMSG study evolves from initiating idea to a mature running protocol, is complex and may vary from trial to trial. Alternative study initiatives are continuously discussed in the steering group and usually penetrated more in detail by designated ad hoc committees. The question to be studied has to be of international scientific interest, but at the same time relevant ‘on the floor’ for the participating local clinicians, and studies should be adapted to the strengths of the group and be in harmony with our guiding principles (see below). Of uttermost importance is that the chosen protocol chairman has a genuine interest in the scientific question to be studied, has a driving and enthusiastic attitude and is (or can be supported by) an experienced ‘clinical trialist’. All new protocols have to be thoroughly penetrated by and discussed with the steering group, the regional coordinators and representatives from all local participating clinics. The time from conception to activation of a study is dependent on several factors, e.g. the experienced relevance of the scientific question, the processing of the necessary paperwork, organisation of monitoring, and negotiating with supporting pharmaceutical companies, but will never fall short of 9 months.
Principles of trial design and performance
During the years with the group, we have found it important – in our hands and in our setting – to adhere to certain principles with regard to design and performance of clinical trials.
The trials should be population based
The road had been paved by several population-based incidence studies in Sweden and Norway, showing the incidence of multiple myeloma to be 5–6 per 100 000 (11–13), and the median age of patients in unselected materials to be 70–72 yr. The national cancer registries in the Nordic countries are very complete and include for multiple myeloma almost 100% of newly diagnosed cases. The low mobility of the population and a healthcare organisation where each individual has a long-lasting association with one hospital allow for high accrual and follow-up rates in multicentre studies. To be able to generalise the trial results to the general patient population we have therefore set as a goal that in studies performed by NMSG, all patients diagnosed during the study period in the recruitment area should be accounted for, whether or not they are included in the trial. It was natural with this intention to invite all hospitals, including small ones, to join the NMSG trials. In the first large protocol (#4/90) on the significance of adding IFN for initial and maintenance therapy 67% of the estimated total number of newly diagnosed myeloma patients were registered in the study and 38% were in fact randomised (13). In a later trial on the impact on survival of high-dose therapy with autologous stem cell support in younger patients (#5/94), 77% of the expected number of patients were registered and as many as 61% were included in the study; in protocol #7/97 the corresponding figures were 78% and 71% (unpublished data). The quality of the performance of the small county hospitals, as compared with major county hospitals and university hospitals, was investigated for the MP vs. MP-IFN trial (15). Neither the accrual rate nor the adherence to the protocol was found to differ significantly between the three hospital categories.
The main trials should be investigator initiated
There was early a consensus in the group that the scientific problems addressed in prospective trials should be defined within the group. While some support from pharmaceutical companies, such as free drug or general financial support to the NMSG organisation or unconditioned research grants have been accepted and found necessary to run the trials, industry has not had any influence on the study protocols. Offers to involve NMSG as a group in trials planned and run by pharmaceutical companies have so far been declined. Recently, this principle caused great delay of an on-going trial aiming to define the role of bisphosphonates in the prophylaxis of skeletal disease in multiple myeloma. When preliminary results of the randomised trial, comparing 90 mg pamidronate i.v. every 4 wk to placebo, were published (16) and NMSG wanted to run a corroborative, placebo-controlled trial of unselected patients and with quality of life as one of the main end-points, no company was interested in supporting such a study. After much delay, a less than ideal protocol (#8/00) comparing 30–90 mg pamidronate i.v. every fourth week was initiated without industrial support. As a positive sign that health authorities recognise the importance of trials that are independent of the pharmaceutical industry, the Norwegian Health Department provides substantial financial support to the Norwegian part of that study. The conviction that the group as such should not take part in pharmaceutical company-initiated studies has not precluded that individual centres ‘belonging’ to NMSG (usually university or large county hospitals) have joined some recent international protocols (erythropoietin, thalidomide and bortezomib).
Comparison with historical controls may sometimes be an acceptable alternative to a randomised study
When NMSG in 1992–93 planned for a randomised study of high-dose melphalan followed by stem cell support vs. traditional oral MP in younger patients, reports of the advantages of the intensive therapy regimen were already circulating among both doctors and patients. After much consideration we had to accept the fact that such a study would have great difficulties to recruit patients and maybe needed to be extended over 6–7 yr. We instead decided that one-armed studies might be better suited to give us what we wanted: rapid introduction of this type of therapy at all the group's university hospitals under controlled conditions, study of the pros and cons of the regimen, including quality of life and health economics, and evaluation of the outcomes against the background of comparable patients from our own historical control material from previous Nordic studies, all with high representativeness. We are convinced that this study design – for the moment – was the best for our group and for our patients, and that it – under certain conditions – may be used as an alternative to randomised studies. This may especially be the case when evaluating new ‘drastic’ therapies, and when one of the tentative arms in the study may be of preference to patients and/or participating physicians.
The trials should include quality of life and cost-utility analyses
The most distressing symptoms of myeloma are skeletal pain, fatigue and reduced physical and role functioning. Altogether, these symptoms cause a substantial reduction of the patients’ quality of life. At the same time, some of the treatment modalities that have been tested and implemented in this disease may themselves cause significant side-effects with a bearing on quality of life. For these reasons, we have considered it important that the endpoints of NMSG trials should not be restricted to response, response duration and overall survival, but also include quality of life. Starting with the MP vs. MP-IFN trial (#4/90) we have used the questionnaire developed by the European Organisation for Research and Treatment of Cancer (EORTC QLQ-C30), which was validated for use with myeloma patients by the NMSG (17, 18). This technique was well suited to demonstrate the significant effect of IFN therapy on the patients’ quality of life (18). In this and later trials, prospective data on resource consumption have also been collected. In collaboration with experts on health economy, a cost utility analysis has been performed for several treatment modalities including IFN (19) and intensive therapy with stem cell support (20). The impact on the disease of therapy with bisphosphonates (#8/00) and with thalidomide (#12/02) is currently under investigation.
The design of the trials should be simple
In order to ensure a high accrual rate and participation also from hospitals without research nurse support, it is important that patient inclusion and follow-up, necessary investigations and completion of Case Report Forms, add as little as possible to the clinical burden of participating physicians. The aim has been to construct protocols in accordance with the NMSG guidelines (see below), stimulating the doctors to follow their patients systematically, with as little additional work as possible. However, one such extra task specified in the protocols of all prospective NMSG trials, has been and is to store samples of biological material from study patients for later investigations.
Trials should take place within the frame of a collaborative network with due informative and educational support
In each region meetings with participating physicians have been arranged at least twice a year, and regular newsletters describing the progress of ongoing trials are distributed. In a study performed by the group in 1995, 99 of 107 physicians taking part in the MP vs. MP-IFN trial (#4/90) were questioned regarding their attitudes to clinical trials (21). The scientific purpose of a trial was considered to be the most important factor for patient accrual, followed by simplicity of the study protocol, ethical quality of the study, ease of communication with the study organisation, and the participation in investigators’ meetings. There was a positive correlation to patient accrual for the physicians’ perception of the importance of a quality of life study in the protocol. Thus, the importance of principles underlying NMSG trials discussed above was clearly confirmed by the collaborating physicians.
The main trials should stimulate spin off projects
The general idea has been that data from our main trials, as well as biological material stored from study patients, should be available for specialised studies. All such studies have to be approved by the Steering Group, but they may be initiated by the study leadership, members of the steering group or by other individuals within NMSG. A number of studies have thus been performed in order to evaluate new prognostic factors (22), and the impact of renal failure (23, 24). The importance of the serum concentration of substances such as syndecan-1, hyaluronan, hepatocyte growth factor, insulin growth factor and osteoprotegerin has also been studied, mainly by the group in Trondheim (25–29), and of serum markers of bone metabolism by the group in Aarhus (30). The quality of autologous stem cell collection and evaluation was investigated in a quality assurance programme tied to the high-dose melphalan trials (31–33).
As previously mentioned, a number of trial design and performance studies were carried out in connection with the MP vs. MP-IFN trial (#4/90). Special interest was focused on defining inclusion and exclusion of patients, on quality of participation from different types of hospitals, and of physicians’ attitudes to participate in clinical trials.
In spin-off projects data from NMSG trials have been used to compare compliance in various quality of life studies (34) and to assess the consistency of the psychometric properties of the various quality of life scales within the EORTC QLQ-C30 questionnaire (35). NMSG has also contributed to the development of an EORTC questionnaire module to be used in quality of life assessment for multiple myeloma patients (36).
Starting with the first large NMSG trials, certain common guidelines were included in the protocols regarding diagnostic criteria, inclusion of patients, treatment schedules and criteria of response and relapse. These were for each study prepared by the responsible protocol committee, and based on their interpretation of available published data and the experiences of our own previous studies.
In 1995 a first draft of ‘NMSG Guidelines for Diagnosis and Treatment of Multiple Myeloma’ was prepared by a selected working party. These guidelines were expanded and included all aspects of patient care, among them patient information and physical training issues. The document was referred for comments to all regional coordinators, discussed with the representatives from local hospitals, and finally approved by the Steering Group and distributed to all hospitals in Nordic countries (Danish, Norwegian and Swedish editions) as a ‘gold standard’.
A revision of the 1995 Guidelines was recently initiated by a working party, distributing chapters of the text among a total of 26 experts within the NMSG organisation. Each of them updated the text in accordance with published data and the experience from our own studies. The results were discussed and consensus conclusions were drawn at two general meetings with the Steering Group. Finally, the guidelines were sent to all regional coordinators for formal approval within their region. A web-based edition in the Scandinavian languages, but with recommendations translated into English, is since 2003 available on the NMSG home page (http://www.nordic-myeloma.org). At present a coordination project is ongoing with the aim to establish common UK Myeloma Forum/NMSG Guidelines, in accordance with published criteria for ‘evidence-based medicine’ (37).
The NMSG is organised as a Nordic enterprise, based on communication in Scandinavian languages and with a moderate recruitment area (in practice around 15 million inhabitants; potential if all Nordic countries fully participated, 24 million). These conditions naturally lead to certain restrictions in our work, but also to advantages: uniform structure of the medical care and regulation systems, good personal knowledge of participating physicians, ease of communication and long-term stability in the patient–physician relationship. We are at present not prepared to give up these advantages, and would like to retain the best of them in the planned European Myeloma Network (EMN) by favouring cooperation between trial groups rather than between individual centres in a common ‘federal’ study organisation. Principles for such intergroup studies have recently been set up by the EORTC (38), which the NMSG easily could accept.
Patient recruitment to the group's initial trials was no problem: the investigated treatments could easily be transferred into clinical practice and many hospitals eagerly seized the opportunity to take part in trials they considered important, thereby also becoming part of an international network with up-to-date clinical and scientific standards. Our intensive treatment protocols also recruited very well, but mainly because they were centred around the university clinics, where physicians have more time for research activities and research nurse support is more readily available. However, in our latest trials recruitment has declined. Several explanations for this development may exist:
The scientific problem, addressed by the study, may seem less relevant or is in conflict with (at least some) international opinion. This may be the case for our pamidronate study (#8/00), where forceful marketing efforts of several substances (oral chlodronate, i.v. pamidronate and zoledronate) have counteracted study recruitment.
The motivation of local physicians to participate in our studies has been hampered by several potential mechanisms: general ‘lack of time’ experienced by all clinicians during later years, caused by increasing demands for efficiency due to economic restrictions of health care; competition between trials and associated spin-off projects (easier when one trial was run at a time); investigator-initiated trials cannot compete with pharmaceutical industry trials with regard to financial support (substantial investigator fees, payment for research nurses, ‘fringe benefits’); multicentre studies are not – with the present system of authorship – considered to give enough author/investigator credit to participating physicians.
New regulations, recently introduced both in our countries and common for the European Union, have made the prerequisites for starting and running investigator-initiated clinical trials more complicated and laborious. To fulfil all aspects of the good clinical practice regulations, especially with regard to monitoring, and the new rules for Ethical Committee approval, needs more time, investigator and staff resources and economic funding than many clinical trial groups today possess.
The performance of NMSG during its first 15 yr has emphasised the need for and advantages of a close cooperation between the Nordic countries in order to improve the general medical care of myeloma patients at all health care levels, to critically evaluate and support the introduction of new therapeutic advances, and to stimulate local research projects. In spite of the small recruitment area, the group has through a high accrual rate been able to perform a number of relevant clinical trials comprising more than 2500 patients, often with a distinct profile, and through them contribute to internationally valid knowledge.
Future prospects and challenges
Independent clinical trial groups are in the future facing a number of challenges.
The need to define and attack the most important scientific question
In recent years a number of new drugs with seemingly high potency in multiple myeloma has been introduced. How these drugs should be optimally used in different phases of the myeloma disorder is not known today, but has to be rapidly evaluated through large, representative collaborative studies. A great step forward for the myeloma patients globally could be achieved if international resources in the coming years were united for this purpose. As all the new drugs will be very expensive, cost effectiveness studies should be an essential part of such efforts.
The need of financial support
Ideally this should come from several sources (research foundations, EU research programmes, non-profit organisations, government, pharmaceutical industry). A substantial improvement would occur if the research councils of the Nordic countries and elsewhere upgraded, and thereby better supported, the really important clinical trials and outcomes research projects. It should also seem pertinent if national or regional health authorities to a greater extent had the ambition to support – after due evaluation – trials with a strong emphasis on health economics; such support has so far been given to our group only for the pamidronate study (#8/00) by the Norwegian Health Department. A further approach would be to plan and run clinical trials in closer connection with, and support from, the two large international myeloma research and patient support organisations, International Myeloma Foundation (IMF) and Multiple Myeloma Research Foundation (MMRF). When it comes to collaboration with the pharmaceutical industry, which for many reasons is important, it would be advantageous if groups with a similar approach as the NMSG could unite and try to define certain guidelines, aiming at transparency of financial agreements and maintenance of scientific independence.
The need to recruit large numbers of patients in a short time
This can be achieved only if two conditions are fulfilled: the group has to have a large enough recruitment area, and there must be incentives for local clinicians to participate in a trial. The NMSG has been able to recruit up to 200 new cases per year of myeloma in ‘younger’ patients (below the age of 65), but has for the moment difficulties when it comes to elderly patients, where small local hospitals have to be more acitively involved. So, while some studies still can be performed within the group's domain, other studies may necessitate cooperation with other myeloma trial organisations. Among incentives important to trial participants are educational spin-offs (as discussed above) and economic support for (partially) financing research nurses. It is also recommended that a more detailed presentation of participants, their contributions and duties within a clinical trial should be included in its final publication (e.g. as an appendix), and that participation defined in this way could be used, besides formal co-authorship, in merit evaluation.
These issues have recently been discussed also by other clinical trial groups (39, 40). One fruitful approach to an improved management of the problems presented above might be the creation of clinical trial group networks, in line with the ongoing efforts of the EMN. This may lead to more potent trial organisations, capable of attracting sufficient financial support, and better recruitment of patients from larger populations The challenges facing the myeloma scientific community both in Europe and in the US with the arrival of new effective treatment alternatives, and an increasing consensus on good clinical trial methodology, have already stimulated a closer cooperation between groups and institutions. Our belief is that this process will be accelerated in coming years.
The Nordic Myeloma Study Group has since 1997 been supported by a grant from the Nordic Cancer Union. Grants supporting the organisation have also been received from Amgen, Celgene, Janssen-Cilag, Roche and Schering-Plough.