Discussion
The panel have made a series of recommendations which define a group of patients in whom the use of biologics is considered appropriate in the treatment of CRSwNP, and a group where use would be considered inappropriate, considering the current evidence base, costs and capacity in the setting of the NHS.
The RAND/UCLA appropriateness methodology is well described and has previously been utilised defining appropriateness criteria for ESS during management of uncomplicated adult CRS and adult recurrent acute rhinosinusitis (1, 11, 12). We have used a similar process to develop BRS guidance on the management of incidental findings in the maxillary sinuses with regard to dental implantation(13), and to develop guidance on the treatment of COVID-19 related loss of smell(14). This exercise or process aims to detect and achieve consensus amongst a group of experts, and is ideally suited to evaluating the appropriateness of use of medical interventions where the evidence base is limited, which consume significant resources or where use remains controversial. Defining appropriate use of biologics in CRSwNP would therefore seem to be well suited to the methodology.
Biologic therapy using Mabs that block the action of interleukins or other targets central to type 2 inflammation now play an important role in the management of difficult-to-treat asthma and many of these treatments have also been shown to be effective in the management of severe CRSwNP. Dupilumab, an anti IL4/13 receptor mab, and omalizumab, an anti-IgE mab, have been shown to achieve significant reductions in polyp size and nasal congestion in large phase 3 studies(15, 16). Both have now been granted FDA and EMA approval for use in patients with CRSwNP, and are currently available for use in the US and selected European countries. Other drugs will likely soon follow, specifically with phase 3 trials completed for mepolizumab(17) and benralizumab, which target IL5.
Although biologicals have been shown to reduce the need for surgical intervention for CRSwNP(17), their high costs and the need for long term treatment mean that this is unlikely to be the most cost-effective treatment across the whole population with CRSwNP, even if superior in terms of long-term symptom control in the difficult-to-treat group. Scangas et al undertook a Markov decision tree cost-effectiveness model over 20 years(18), and found, based on US costs (which may not be applicable in all healthcare setting), that a strategy of sinus surgery cost circa $50,000 producing 9.80 QALYs while dupilumab treatment costs $535,000 but produced 8.95 QALYS. Surgery was more cost-effective regardless of the frequency of revision surgery. Similarly, in asthma, although the efficacy of biologic therapy is well established, none of the currently available drugs have been found to be cost-effective(19).
Currently, no biologic treatments have been approved by NICE for the treatment of patients with CRSwNP. NICE considers evidence of effectiveness of new interventions, but applies a standard threshold range with an upper limit of £20-30,000 per QALY and a budget impact test, where drugs that cost more than £20million in any one of their first three years of use trigger commercial discussions to mitigate the impact on the wider NHS, or further restrictions on usage (www.nice.org.uk). Using the data from Scangas et al, with a £43,500 cost per QALY at current currency conversion rates, it is unlikely that the NICE threshold could be met if biologics were prescribed to all patients with CRSwNP.
However, the panel’s recommendations identify a group who are more likely to fail to achieve long term benefit from conventional treatment pathways and to undergo repeated interventions with higher associated healthcare resource utilisation. A recent study has shown that patients with a history of previous surgery are twice as likely to need further revision surgery, or those with a history of N-ERD five times more likely to require further revision when undergoing endoscopic sinus surgery(20). Therefore, as the patients identified by the panel have higher direct costs over their lifetime than other patients with CRSwNP, use of biologics in this subgroup are more likely to be cost-effective. Patients with higher rates of revision surgery also likely derive less symptomatic benefit from conventional treatment pathways; indeed a recent study found that 43% of patients’ symptoms were uncontrolled after sinus surgery and this again was more common after revision surgery and in the setting of N-ERD(21). Our criteria will also, therefore, help to identify patients for who current treatments are likely unsuccessful in achieving adequate disease control, and where biologics may offer the only option that may achieve long term disease control. We, therefore, hope that NICE and other organisations will consider our criteria to define a group for whom biologics should be approved even if the usual threshold limit per QALY cannot be met, and not seek to impose further restrictions on usage over and above those defined by the panel (Fig 1).
The criteria used by the panel in reaching consensus where biologics are considered appropriate define a smaller group of eligible patients than either the EPOS(8) or EUFOREA(9) criteria for patient selection for biologic therapy. The scenarios where the panel felt that use of a biologic was uncertain included patients with at least one previous surgery, many of whom would meet the EPOS criteria for biologics, although it should be noted that the BRS set a lower threshold of symptom severity at baseline, measured using the SNOT-22 or VAS. Many of the scenarios where use of biologics in the NHS is currently considered ’uncertain’ by the BRS panel represent patients who would have been eligible for recruitment to published trials demonstrating the effectiveness of biologics, and within the trials achieved significant improvements in nasal polyp score and quality of life on biologics. Therefore, it is important to state that the BRS recommendations are not intended to predict response to treatment to biologics, but only to define a group where the panel felt use was appropriate given the financial restrictions within the NHS at the current time. It is very likely that recommendations for the scenarios rated ‘uncertain’ will change if the relative cost of treatment is reduced in the future.
Panellists recommended that biologic treatments for CRSwNP, if approved for use within the NHS, should be delivered within centres of excellence co-located with difficult to treat asthma, where there is pre-existing experience of treatment with biologics. Centres should also be able to offer a range of other treatments. This will also have the likely impact that patients with difficult to treat CRSwNP, for example those with N-ERD, are directed to specialist centres that can ensure that surgery is performed optimally and that patients are considered for adjunctive treatments such as post-operative desensitisation, thus reducing the need for biologics. Indeed, a recent paper has shown that more extensive surgery in patients with CRSwNP was associated with lower rates of revision surgery(20); creation of specialist centres will, therefore, likely improve outcomes from other interventions such as sinus surgery. Panellists commented that in future, the extent of previous surgery and the interval between previous surgery may also be considered in the decision making process, and perhaps patients who have only had more limited surgery be considered for revision surgery before biological therapies.
While the panel made recommendations for use in AFRS, it was noted that clear evidence of efficacy on trials was needed before these should be implemented. The panel also agreed that there was no indication for use in patients with CRS without nasal polyps in the absence of evidence in this group
Once biologic therapies are initiated, the response to treatment must be assessed to determine if a patient should continue treatment; how this should be determined is beyond the remit of the current study. Further limitations of these current recommendations are that biomarkers, such as eosinophil levels in blood or tissue, were not considered in the clinical scenarios, as the panel felt that there is insufficient evidence to determine how these aid selection or predict response to treatment. Biologics were considered as a collective intervention but in reality different biologics, defined by their inflammatory target, differ in terms of costs and effectiveness. The impact of COVID-19 on healthcare delivery in the NHS at the current time was not considered – this may favour biologics over surgery, for example, if demand for surgery greatly exceeds capacity. Finally, we have not included patient preferences for different treatments into our recommendation but these of course play an important role in the final decision making with regards to use if biologics become available within the UK.