The Drug Controller General of India (DCGI), acting on a recommendation from the Subject Expert Committee (SEC) approved an experimental biologic therapy manufactured by Biocon Ltd., one of India’s largest pharmaceutical companies. The approval was based on data from a phase II clinical study on a cohort of 30 patients, and on July 10 waived the requirement to establish the benefit-risk profile for this therapy to treat COVID-19 through a properly designed and conducted phase III clinical study.
This experimental therapeutic candidate, Itolizumab, is an anti-CD6 monoclonal antibody that the drug regulator has previously approved to treat psoriasis.
The manufacturer subsequently promoted this approval on social media and television interviews. Spokespeople for the drug manufacturer used the adjectives “breakthrough drug”, “compelling outcome”, “life-saving” and “unique” to tout the efficacy of this experimental drug candidate during their promotion.
The Indian Council of Medical Research (ICMR) disputed some of these statements made by Biocon in a press conference, where its director general Balram Bhargava said that there was no evidence from clinical studies to show that this drug reduced mortality amongst severely ill COVID-19 patients. Subsequently, the National Task Force on COVID-19 refused to include this drug in the protocol for treating the disease. It was reported that the task force wanted to see more evidence of the drug’s effectiveness in treating COVID-19.
This brings us to the following question: how did the DCGI approve this experimental therapy in the first place? The regulator makes available very little information about the process it follows in granting such approvals. Its decisions are always opaque.
Unfortunately, there is yet no peer-reviewed publication that describes or supports the claims being made by Biocon. Instead, the company has been bolstering its marketing efforts with the help of conversations with ‘key opinion leaders’ who support this drug’s approval. Based on what has been disclosed in these conversations, we try to explain whether the company’s claims are reasonably supported by the information that has been disclosed.
Study design analysis
One way to understand the robustness of a clinical study is to examine its fragility index. This metric indicates how many patients will be required to lose statistical significance in attaining the primary outcome – a key metric for granting approval. The higher this measure, the more robust the findings of the study are compared to random chance. The fragility index for Biocon’s phase II study was 1. The median for this metric across the industry is 8. This means just one patient with different mortality in the control group of the phase II study would suffice take to reduce the study’s statistical significance.
Based on what the company has disclosed during its marketing campaigns, one patient who was 73 years old and was enrolled in the control group died during the study. This patient was an outlier in terms of age relative to the general trial population – and so the death of this patient could easily have changed the statistical significance of the primary outcome, mortality.
In addition, while the number of patients with comorbidities was higher in the treatment group, the number of patients already on ‘non-invasive ventilation’ was higher in the control arm than in the treatment arm: 40% (4/10) versus 20% (4/20). Patients already on respiratory support via non-invasive ventilation are more likely to die, and therefore have a poor outcome for the study. A small cohort of patients is more susceptible to such a baseline imbalance. While this doesn’t invalidate the results of this study, it does raise a question of bias: the control arm, to which the treatment arm is compared, had a higher number of patients who were already at risk of a worse outcome.
Similarly, the last observation carried forward imputation – a statistical method to extrapolate missing study observations – creates a bias in favour of the treatment for secondary endpoints. This is because we don’t know if patients who improved transiently with the treatment actually sustained the benefit of the drug candidate through the entire follow-up period. The lack of a long-term follow-up risks missing study observations for patients in the treatment arm who may have worsened on the secondary outcomes, which would have been measured after a temporary improvement. While this is an industry acceptable method, for studies of small duration like the one in question here, it does create a bias in favour of the drug candidate being evaluated.
We should also remember that this was an open-label trial, so the effect of bias due to lack of blinding is another key factor.
Further, the company has disclosed that while 22 patients were randomised to Itolizumab, 20 received treatment during the study (after giving informed consent) and that two withdrew consent. We need to know the study outcomes for these two patients because industry norms for clinical trial analysis mandate analysis on the “intention to treat” principle. That is, once a study participant has been randomised into the study, their outcomes should be part of the analysis. This is important because it is common in the real world to preferentially administer experimental drug candidates to younger patients because they have better chances of survival.
In the recent RECOVERY trial, for example, younger patients were more likely to receive mechanical ventilation than the elderly. Therefore, the study analysis had to account for this difference in the age of the cohort. Analysing study observations on the basis of intention to treat rather than completion of protocol removes this bias.
Taken together, all this data points to the fact that the study’s design and therefore its findings that were advertised were less than robust. The epidemiologist and public health physician Dr Jammi Nagaraj Rao has explained the true size of the patient cohort needed if the study were to meet its primary endpoints – reduction in mortality rate – or to have compelling data. To demonstrate a 40% reduction in risk of death from the, baseline for example, the number of patients that needed to have been enrolled in this study is more than 15-times as many (576).
To be clear, there are instances in which drugs in oncology are provided accelerated approval on surrogate markers alone (not mortality, as was the case here). But in all such cases, a confirmatory phase III study is mandatory. In the case of Itolizumab, DCGI’s decision to waive a phase III study by Biocon is baffling.
All of this brings us back to the core issue plaguing drug regulation and its administration in India: transparency. We have no way to ascertain if the approval by DCGI to the company was on good faith, given the phase II study’s analysis above. And why would the National Task force on COVID-19 refuse to include this drug in its protocol to treat COVID-19 after it had been approved by the DCGI? Isn’t it strange that two government bodies come to such orthogonal conclusions based on the same set of data?
In 2012, questionable approvals by the DCGI were followed by the 59th report of the Parliamentary Standing Committee on Health severely dressing down the Central Drugs Standard Control Organisation. More recently, after stalling us for over two years, the DCGI produced a copy of the report of the Mohapatra Committee the day before a hearing at the Central Information Commission, constituted to inquire questionable approvals that the Parliamentary Standing Committee had pointed out. This report singled out three former DCGIs by name for granting approvals sans scientific or clinical rationale.
Sadly, it may be too late for the actions of the National Task Force on COVID-19 vis-à-vis Itolizumab. Marketing and promotional campaigns led by the manufacturer would have already capitalised on this suspect approval, resulting in a potential windfall for the company.
Anupam Kumar Singh, MD is an assistant professor in the department of internal medicine in Santosh Medical College, Ghaziabad. Dinesh Thakur, MS is a public health activist.