The Implementation of Intensity-modulated Radiotherapy in the UK

Policy Background 

The NRAG report [3] and the Cancer Reform Strategy [5] both recognised the substantial underprovision of radiotherapy in England and that correcting this would be an important step in improving cancer outcomes. An earlier European Society for Therapeutic Radiology and Oncology report showed that in 2004 England had only 50% of the radiotherapy capacity required by its population [6]: this was broadly confirmed by the NRAG report. Other countries, such as France, Belgium and Sweden, had 90% of the radiotherapy capacity modelled for their populations [6].

There is no central funding for radiotherapy in England: the levers for change are local need and national policy. Commissioners need to understand the importance and priority of this service and its role in cure. Scotland and Wales now have different health systems and have undertaken reviews of their own radiotherapy provision [7], [8]. Increasing cancer incidence in an ageing population [9] will be a major driver of increased demand for radiotherapy. Unless action is taken, access rates will fall further and waits for treatment will start to rise [1], [2].

The English NRIG was established by the National Cancer Director in June 2008 with a remit to provide an oversight role, as clinical teams and commissioning bodies develop local responses to the NRAG report [3]. The terms of reference include:

NRIG draws its membership from professional bodies and colleges, the Department of Health, the National Cancer Action Team, cancer networks and specialist commissioning groups [10]. It has subgroups on technology and work force and supports work on contracting frameworks and the development of a tariff for 2012. It is also involved in the development of the national radiotherapy dataset, cancer waits and the provision of proton therapy. The group will commission and develop national work in support of local teams, particularly around known areas of difficulty.

Cost-effectiveness of Radiotherapy 

Radiotherapy is a relatively cheap and effective treatment that consumes a low proportion of the total cancer budget, amounting to 5% in both England [5] and Sweden [11]. It is wrongly perceived as an expensive service. This is because linear accelerators now cost approximately £1.5 million and need to be sited in a bunker costing over £1.5 million. There are also staff costs, which, as in other health-related activity, dominate long-term financial planning, accounting for 54% of the costs of radiotherapy [12]. However, surgery, for example, also has substantial capital and staff costs and it is important to examine the costs and cost-effectiveness of the modality. An international review published in 2008 concluded that radiotherapy offered very good value for money, with a 21 fraction course of treatment costing €3239 ± €566 [12]. The real increase in cost over the last 15 years was estimated to be 5.5% annually [12]. An earlier report estimated the cost per life year gained (excluding skin cancer) in 1995 as $7186 [13], substantially less than the cut-off of £20 000–30 000 per quality adjusted life year used by the National Institute for Health and Clinical Excellence (NICE).

Evidence Base for Intensity-modulated Radiotherapy 

Two major reviews have been published on the use of IMRT [14], [15]. These show that a large number of patients have been accrued into clinical trials, but many of these have not yet been reported. The timescales and power are such that most aim to study acute or late side-effects. The published data are summarised elsewhere in this journal [15], but, in brief, in head and neck cancer inverse-planned IMRT has been shown to produce superior plans, and a reduction in xerostomia (dry mouth) has been proven in three randomised clinical trials, including one carried out in the UK. In breast cancer, there is evidence from two randomised trials that forward-planned tangential breast IMRT can be used to reduce the toxicity caused by the inherent dose inhomogeneity of conventional breast radiotherapy. The three-dimensional assessment of the dose distribution should now be a standard procedure, so that IMRT can be offered to patients who would otherwise be exposed to a high degree of dose inhomogeneity and, thus, an increased risk of long-term late effects, such as breast pain and distortion. In prostate cancer, trials have shown that a higher dose will lead to higher rates of local control and cure. Better conformation of dose to the target reduces toxicity and late rectal toxicity can be reduced by inverse-planned IMRT.

Demand for Intensity-modulated Radiotherapy 

There will be clinical scenarios, such as locally advanced head and neck cancer, where a clear advantage has been shown for many patients; there will be anatomical sites where IMRT should be implemented routinely, e.g. improved dose distribution for some cases of breast cancer and dose escalation for prostate cancer; and there will be sites where individual difficulties in planning determine the choice of IMRT, e.g. certain brain tumours and tumours close to the spinal cord.

Table 1 uses data from a survey of radiotherapy usage in 2005 by the Royal College of Radiologists [2]. IMRT is currently only relevant to the 53% of patients being treated radically for cure and they received 80% of all fractions prescribed. Onto these data have been mapped indications for IMRT and the proportion of patients likely to benefit. The proportion of radical fractions that should be given as IMRT has then been calculated. Breast cancer is identified separately as it can be treated with a less resource intensive forward-planned solution. As the disease is so common and as radiotherapy has a pivotal role in its management, this would account for about 9% of all radiotherapy fractions. Inverse-planned IMRT should be used for a variety of indications and the disease sites identified for early implementation amount to about 25% of all radiotherapy fractions; prostate cancer dominates these indications because it is so common, but it should be noted that for some patients a forward-planned class solution seems to be quite satisfactory. Overall, it is estimated that about 33% of radical fractions should be delivered using IMRT. These are global national estimates and demand should be based on local data developed with clinicians. There is a marked contrast with a survey conducted in 2008 that showed that overall only 2% of patients in the UK were treated with inverse-planned IMRT and 11% with forward-planned IMRT [16].

Table 1. Estimate of the percentage of radically treated patients who would probably benefit from intensity-modulated radiotherapy (IMRT) and the consequent proportion of radical fractions that should be delivered as IMRT

	Tumour site	Percentage of radical radiotherapy fractions	Percentage of patients who would benefit	Percentage of radical fractions as IMRT
				Forward planned	Inverse planned
	Breast	30	30	9	-
	Prostate	16	80		13
	Gynaecological	5	20		1
	Head and neck	8	80		6
	Central nervous system	3	60		2
	Other sites	10	20		2
			Total	9	24
			Grand total		33

Cost-effectiveness of Intensity-modulated Radiotherapy 

The additional treatment costs of IMRT have been calculated in a single study, in which it was claimed that, compared with conformal radiotherapy, costs increased by about 30% because of slower delivery. However, there was no clearly defined comparator [17]. A study from the UK assessed treatment times using the basic treatment equivalent (BTE) model and showed that IMRT delivery using TomoTherapy is actually faster than conformal radiotherapy for head and neck and prostate cancer [18]. An earlier linear accelerator-based study had shown that treatment time changed little, although there was an increase in time for planning and quality assurance [19].

To estimate incremental cost-effectiveness, gains must be quantitated in terms of quality adjusted life years. There have been two studies of the cost-effectiveness of IMRT, both in prostate cancer. An American study was based on the four-fold difference in reimbursement rates for IMRT ($42 000) compared with three-dimensional conformal treatment ($10 000) [20]. There was disagreement between members of the expert panel and no definite conclusion was reached [20]. A recent UK study reviewed 13 non-randomised studies [21]. The authors estimated the additional staff costs of providing IMRT at £1100 per case. The data support the theory that higher doses up to 81Gy improve biochemical control, concurring with earlier dose escalation studies using conformal radiotherapy [22]. The authors modelled progression to clinical failure and death from the reported surrogate outcome of biochemical failure. If higher doses do indeed improve overall survival, then IMRT should be cost-effective [21]. At the minimum dose of 74Gy recently recommended by NICE [23], conformal radiotherapy is safe and the cost benefit depends on the magnitude of the reduction in gastrointestinal toxicity that can be achieved by the use of IMRT compared with conformal treatment. The authors concluded that the magnitude of the benefit and its cost are highly uncertain and this in turn makes cost-effectiveness uncertain [21]. Further data are therefore required.

NICE undertook an initial scoping consideration of IMRT. This included discussion with professional bodies, clinical leads and national policy leads. Their conclusion was that this is not a new treatment (having evolved from conformal radiotherapy); NICE therefore recognised that it was appropriate for guidance to be developed by the professional groups. Also, as four-dimensional adaptive radiotherapy, including IMRT, was a recommendation within both the Cancer Reform Strategy and the NRAG report, this was in fact national policy, as de facto agreement had already been provided by government. It should therefore be taken forward in the National Health Service, led by the National Cancer Action Team and the NRIG technology subgroup, and offered to those patients who may benefit from it.

Intensity-modulated Radiotherapy Training Programme 

In England, all replacement and new linear accelerators purchased through nationally funded initiatives are IMRT capable. As a result, 97% of all linear accelerators in England have the technical capability and the software licences to deliver IMRT. Nevertheless, radiotherapy services in England are at different stages of developing and delivering IMRT. A few have a comprehensive IMRT portfolio and are confident in delivery of this treatment modality; some have yet to begin [16], [24]. Many departments wish to implement these developments to improve the care of their patients, but have identified obstacles, including shortage of staff and lack of funding, together with a lack of training and confidence in implementation [16].

Forward-planned IMRT for breast cancer patients is a relatively simple development based on an assessment of three-dimensional breast plans for compliance with ICRU recommendations. Inverse-planned IMRT is a more complex undertaking and NRIG has developed a tender specification to provide implementation training and support to all those departments wishing to develop IMRT. The use of specific equipment is taught by manufacturers; implementation training assists departments to develop the processes required (such as pathway and protocol creation) and builds confidence when treating the first patients. This approach has proven effective in a trial setting where centres have been assisted to undertake their first IMRT treatments in head and neck cancer [25]. A similar developmental approach was used in the introduction of conformal radiotherapy for prostate cancer [26] and is being used for IMRT in the CHHiP trial [27].

It is expected that at least one department in each cancer network requesting this support will have completed their training and have clinically implemented IMRT locally by 2012. Some of the main points of the programme are summarised here. It is a prerequisite for radiotherapy services to enter into this training programme that:

In summary, all practical obstacles to the routine delivery of IMRT should have been removed before implementation training commences.

The training programme will consist of the following elements:

Peer Review Standards and Intensity-modulated Radiotherapy 

The national cancer peer review measures have been an important driver for change. The standards for radiotherapy have been extensively revised and have recently been published after consultation [28]. They include standards for IMRT based on recommendations by the professional bodies [29]. The following bullet points summarise some of the new standards that apply to IMRT:

Members of the team should have completed training judged satisfactory by the peer reviewers.

Patient Information and Consent 

IMRT is a technique that can deliver improved dose distributions compared with conventional techniques. The General Medical Council requires that patients are told of all options for their treatment [30] and it is now clear that IMRT should be discussed as an option for patients with advanced head and neck cancer, selected cases of breast cancer and many cases of prostate cancer. There are a wide range of other indications and indeed some patients may be essentially untreatable without an IMRT solution. This technology should be available to all clinical oncologists when deciding the optimum treatment for their patients and its role should be discussed with those for whom it may be beneficial. Thus, as a minimum, every cancer network should provide at least one IMRT service.

Research 

One of the aims of expanding IMRT capability is to increase trial entry, not just to provide further evidence that current improvements in dose distribution improve patient outcomes in a range of clinical scenarios [14], [15], but also to ensure that the UK radiotherapy research community is in a position to evaluate novel techniques that should deliver higher doses safely and may thus increase cure rates [31].The testing of radiotherapy in combination with cytoxics and novel agents will require techniques that are technically optimal.

Summary 

This editorial lays out the rationale for the implementation of IMRT as the standard of care for many radiotherapy patients and a proposed mechanism to help departments to implement it.

At least one third of breast cancer patients should be offered the relatively simple technique of forward-planned IMRT to improve dose distribution and decrease the risk of distressing long-term side-effects in the conserved breast: this will apply to about 9% of all radical radiotherapy fractions delivered, as the disease is so common. A range of other indications (e.g. head and neck cancer) will account for the use of inverse-planned IMRT for about 24% of all radical radiotherapy fractions, making a total with breast cancer of about 33% of radical radiotherapy fractions. The aim, and the challenge, is to achieve this by 2012.