Phase II for immunomodulators: what hypotheses should we test?

In pharmaphorum's inflammation and immunology themed month, Alexander Gray of IDEA pharma explores the similarities between oncology and immunology.

As with oncology, the broad church of 'inflammation and immune modulation' has proven attractive for Pharma companies, who have been enticed by the combination of significant unmet need, secondary / tertiary care markets, and potential for relatively high pricing.

As has occurred in oncology, we have come a long way in the last few decades. The launch of the anti-TNFs over 10 years ago heralded the start of a more targeted approach to the treatment of immune-related disorders than the more generalised, small molecule immunosuppressive strategies used previously. This has led to significant enhancements in patient care: not only a greater impact on symptoms, but disease modification, and enhanced quality of life with targeted agents delivering superior toxicity profiles.

However, the industry has tended to head rapidly towards Phase II studies in higher-incidence inflammatory conditions such as rheumatoid arthritis, with few tackling lower incidence conditions such as systemic lupus erythematosus (SLE), despite significant unmet need. In their article 'Diagnosing the decline in pharmaceutical R&D efficiency', Scannell and colleagues discuss the "narrow clinical search problem": the tendency in pharma to ask narrow questions in Phase II studies.1 Of course asking specific questions has upsides: small studies, and potentially less chance of failure when Phase III is initiated. The downsides though are more concerning: primarily the tendency for mechanisms to 'fail' because the question that was asked was too specific (disease rather than mechanism), and the sample size too small to understand in whom the mechanism might work, rather than the average poorer effect across a population. In immunology, the approach to Phase II may also have been coloured by the trailblazing in RA, where over a decade in research has failed to provide prospective biomarkers for the benefit of anti-TNFs.

One similarity between oncology and immunology is the complex nature of cell pathways 'fuelling the fire' of the disease. In cancer, there are a multitude of cell growth pathways that can be aberrant at different stages of the disease. In immunology, the problem is not dissimilar but is more 'parallel' rather than 'series': multiple pathways that drive the aberrant inflammation. An example of this is IL-4 and IL-13 in asthma: it may well be the case that blocking both will be better than either alone.2 In addition, such pathways often have a role across a range of inflammatory conditions. 'Anti T-cell' or 'anti B-cell' monoclonal antibodies, or small molecules directed against targets such as Janus kinase (JAK) have widespread effects on the immune system overall, and therefore could impact a host of immune-related disorders.

What we can learn from this marvel of evolutionary biology is that:

1) We are rarely targeting immunologic diseases per se: we are targeting pathways

2) The value of inhibiting these pathways is unlikely to be the same in different diseases or their manifestations

3) The value of inhibiting these pathways is unlikely to be the same in different patients, or even at different times: we therefore should likely stop asking questions such as 'is RA a T-cell or a B-cell disease' and start asking 'what are the drivers of disease in this patient'

So, what other approaches might exist? Phase IIa, and perhaps even Ib, could provide an opportunity to look more broadly across disease states, and more deeply, within them. A high degree of objective response in a proportion of patients could then be explored more deeply. Such an approach also affords the opportunity for the use of surrogate biomarkers (an increasing feature of approvals in this area3) to examine in whom and even when mechanisms provide meaningful effect: looking for correlations between clinical benefit and cellular or molecular outcomes. A good example of this is the work examining the benefit profile of anti-IL-5 monoclonal antibodies in asthma: blood and sputum eosinophils, and serum periostin, appear to have value in predicting response to this mechanism.4

Another important opportunity that lies with such testing is better delineation of the disease states being examined. Inhibiting mechanisms such as IL-5 in asthma has already provided great insights into the pathogenesis of the disease: IL-5 trap appears to have a much greater benefit on reducing exacerbations than it does on improving lung function, for example.5 This observation matters both at a disease level, and of course for planning future studies, and their choice of endpoints.

"I have a strong suspicion that the management of complex immunologic disorders is going to become more like oncology: targeting of mechanisms, not diseases."

Another great observation from these studies, in parallel with other research, is that airways disease (asthma and COPD) is a spectrum, and that eosinophils play more of a role in some variants than others. I doubt it will be long before we acknowledge that the traditional boundaries of these diseases will start to break down as we realise that classifying patients as 'eosinophilic' or 'neutrophilic' may have more value for directing therapy than 'asthmatic', 'bronchitic', or 'emphysematous'.

I have a strong suspicion that the management of complex immunologic disorders is going to become more like oncology: targeting of mechanisms, not diseases. One (but not the only) trend in oncology, which I have little doubt will continue, is segmentation of patients both before and during treatment, on the basis of which mechanisms are driving their disease. The value of doing broader-based, early-phase clinical studies in immunology, is that we will uncover the role of different driver pathways across diseases, and understand better how to target therapeutics to patients, not a diagnosis.

The take-home message for the industry is that they need to be more open-minded about the purpose of early- and mid-phase clinical studies, exploring not only whether a mechanism has value, but on which parameters, in whom, and how. The tendency for pharma to look for their keys under single lamp-posts, and to move on (backwards or sideways) if they can't find them, is unlikely to deliver better solutions for patients and physicians, or product successes for pharma.

References

1. Scannell JW, et al. Diagnosing the decline in pharmaceutical R&D efficiency. Nature Reviews Drug Discovery 2012; 11: 191–200.

2. Wenzel S, et al. Dupilumab in persistent asthma with elevated eosinophil levels. New Engl J Med 2013; 368: 2455–66.

3. Downing NS, et al. Clinical trial evidence supporting FDA approval of novel therapeutic agents, 2005–2012. JAMA 2014; 311: 368–377.

4. Jia G et al. Periostin is a systemic biomarker of eosinophilic airway inflammation in asthmatic patients. J Allergy Clin Immunol 2012; 130: 647–654.

5. Pavord ID et al. Mepolizumab for severe eosinophilic asthma (DREAM): a multicentre, double-blind, placebo-controlled trial. Lancet 2012; 380: 651– 659.

About the author:

Alexander Gray is Partner and Medical Director at IDEA Pharma.

Alex is a pharmaceutical physician working as an external consultant at the clinical-commercial interface on Global, European and U.S. projects, typically for top ten pharmaceutical companies. In his 17 years in industry, he has consulted on clinical trials strategy development and alignment with commercial brand vision, brand positioning development, product portfolio profiling and strategy development, brand strategy and operational plan development. In his spare time he is a fan of Baroque and rock music, and builder of cars.

He can be emailed at alex.gray@ideapharma.com

Closing thought: will the management of complex immunologic disorders become more like oncology?