Sunday, June 20, 2021

Early Phase Trial to Find Maximal Tolerated Dose (MTD) - 3+3, CRM, and BOIN Designs

In early-phase clinical trials, determining the dose range and therapeutic window is critical. The purpose of the early-phase studies may just be to identify the maximum tolerated dose or maximum tolerable dose. 

Definition of maximum tolerated dose (MTD)
The highest dose of a drug or treatment that does not cause unacceptable side effects. The maximum tolerated dose is determined in clinical trials by testing increasing doses on different groups of people until the highest dose with acceptable side effects is found. Also called MTD.
The studies for identifying the MTD are usually designed as a dose-escalation study and the dose-escalation study is defined as:
A study that determines the best dose of a new drug or treatment. In a dose-escalation study, the dose of the test drug is increased a little at a time in different groups of people (also called cohort) until the highest dose that does not cause harmful side effects is found. A dose-escalation study may also measure ways that the drug is used by the body and is often done as part of a phase I clinical trial. These trials usually include a small number of patients and may include healthy volunteers.

In dose-escalation studies, within each dose cohort, a placebo group can be included even though the majority of the dose-escalation studies for MTD are designed without placebo controls.

Identifying MTD is based on the number of dose-limiting toxicities (DTLs)that are observed in each dose cohort. DTLs are defined as: 

side effects of a drug or other treatment that are serious enough to prevent an increase in dose or level of that treatment.

In practice, DTLs are often defined as grade 3 or above adverse events according to Common Terminology Criteria for Adverse Events (CTCAEs) especially in the oncology area even though other customer-defined criteria for DTLs may be used in non-oncology areas. 

Clinical trials to identify the MTD are generally needed for phase I studies directly conducted in patients, not healthy volunteers. Areas that the phase I studies are conducted in patients, not healthy volunteers, include oncology drugs, drugs in severe diseases such as AIDS, Sepsis, ARDS, etc., the gene and cell therapies, human-plasma derived products.

There are different types of clinical trial designs for identifying the MTD. The commonly used designs are 3+3 design, Continuous Reassessment Method (CRM), and Bayesian Optimal INterval design (BOIN). 

3+3 Design was discussed in an early post Phase I Dose Escalation Study Design: "3 + 3 Design". It is a straightforward rule-based method and requires no statistical calculations. 3+3 design is the most frequently used method for identifying the MTD. 

The CRM is a model-based design for phase I trials, which aims to find the maximum tolerated dose (MTD) of a new therapy. The CRM has been shown to be more accurate in targeting the MTD than traditional rule-based approaches such as the 3 + 3 design. With CRM design, statistical inferences on the model parameter(s) need to be made using likelihood-based or Bayesian approaches and DLT probability at each dose needs to be estimated. The patient is assigned to the next dose level based on the probability of patients with DLTs at the current dose level. The toxicity risk of other dose levels is based on accrued data, which improves trial efficiency. 

Following articles or videos provided a great introduction/reference about the CRM method: 

The BOIN design shares the simplicity of the 3+3 design, which makes the decision of dose escalation/de-escalation by comparing the observed DLT rate with 0/3, 1/3, 2/3, 0/6, 1/6, and 2/6. The BOIN design makes the decision by comparing with two fixed boundaries, λe and λd, which is arguably even simpler.



BOIN design are described and explained in the following article and video:
Software for Sample Size Calculation for Phase I MTD Finding Studies:
  • trialdesign.org is a website developed and maintained by a research team at MD Anderson Cancer and it contains the literature and software for phase I designs including CRM and BOIN. 
Additional Videos: 

Saturday, June 19, 2021

About Controversial Approval of Biogen's Alzheimer Drug

Two weeks ago, the US Food and Drug Administration (FDA) approved aducanumab (brand name Aduhelm) as a treatment for Alzheimer's disease -- a historic decision not because it addresses the longstanding unmet medical need for a safe and effective cure of a devastating disease that affects nearly 6 million Americans, but because of the unprecedented irregularities of the agency's actions, undermining its mission to protect public health and ensure the "safety, efficacy, and security" of treatments made available in the United States. 

The winner is obviously the drug developer, Biogen and its collaborator Eisai. They probably never thought that FDA would be so collaborative and more desired to approve aducanumab than the sponsors themselves. They rescued a drug that had been declared 'unlikely' to work (futility) just two years ago. They got an unlimited label for all Alzheimer patients (beyond the early Alzheimer patients that were studied in their clinical trials). They can decide on the drug price whatever they want because there is no price control in the US once the drug is approved by the US FDA. They have at least 9 years to complete the post-marketing confirmatory study. There are no incentives for them to complete this confirmatory study as early as possible. The longer the study takes, the more time they can make the money from a drug with unproven efficacy. 

The losers include a long list: 
  • FDA - loses its credibility
  • Alzheimer's patients - are given false hope and may end up taking 'snake oil' for many years down the road
  • Patient Advocacy Group - Alzheimer's Association was unhappy with Biogen's $56,000/year/patient price tag. 
  • Medicare/Medicaid/Insurance Companies - extremely high cost associated with Aduhelm ($56,000/year/patient) and the broad label for Aduhelm can cost them a lot of money
  • FDA Adcom Committee - insulted by FDA's decision to approve even though the Adcom voted overwhelmingly against the approval
  • Regulatory science - FDA has touted for years about the regulatory science and the strict rules to be followed for drug approval - these rules are not followed by the FDA - what can you do?
  • FDA statisticians - It is clear that the FDA statistical reviewers had their dissenting opinions and questioned the data / results from two pivotal studies that were prematurely discontinued due to futility. Statisticians' opinions were overruled. 
......

Usually, approval like this will be heralded as historical and celebrated by all parties - not this time for aducanumab approval. The reactions are overwhelmingly negative. Here is a list of articles discussing the controversial approval from different angles.  
In approving Biogen's aducanumab, the boundaries between the FDA (as a regulator) and the sponsor (as a drug developer) were crossed. In the drug development field, the sponsor will try everything to exaggerate the benefit and minimize the side effects while FDA will need to be on the conservative side, tamper down the expectations, prevent the manipulation of the data and biases in data analyses,...  In the aducanumab case, FDA is determined to approving the drug no matter what and no matter whether the data/ results from clinical trials have demonstrated "Substantial Evidence of Effectiveness". FDA retrospectively find a regulatory pathway (accelerated approval pathway) for approval. In doing so, FDA failed to stand by the standards it established and both regulators and sponsors had followed.

In the drug development field, pre-specification is critical. The regulatory pathway, the number of clinical trials for clinical development program, the clinical trial design, study endpoints, and statistical analysis plan have to be discussed and agreed upon with FDA. As Eli Lilly's CEO said that in drug development, "where the gold standard for approval is you call your shot, and then you hit your shot, like Babe Ruth pointing at the left-field and then hitting his home run there." The sub-group analyses and post-hoc analyses are for hypothesis-generating and can not be used to support the regulatory approval. In Biogen's case, it is obvious that an additional clinical trial is needed before the approval. By switching to the accelerated approval pathway, FDA essentially agreed that the pivotal studies with cognitive and function measures provided insufficient evidence for approval and they had to retrofit to find accelerated approval that is based on the biomarker (amyloid).

In a letter from FDA to AdCom about switching to the accelerated approval pathway, Dr. Billy Dunn said this: 
Following the advisory committee meeting, further discussion within FDA considered the uncertainty introduced by the conflicting results of Study 302 and Study 301 and the committee’s discussion of that uncertainty. Our discussions raised further consideration of the accelerated approval pathway; a topic discussed earlier in the development program but not directly discussed during the advisory committee meeting given the focus at that meeting on the evidence of clinical benefit. As you may be aware, the accelerated approval pathway is for drugs to treat serious diseases that are expected to provide a meaningful advantage over available therapy, but where there is residual uncertainty regarding the drug’s ultimate clinical benefit. To be approved under this pathway, there must be substantial evidence of the drug’s effectiveness on a surrogate endpoint—usually an endpoint that reflects the underlying disease pathology (accelerated approval can also use an intermediate clinical endpoint). An effect on this surrogate endpoint must be shown to be reasonably likely to predict clinical benefit. We concluded that these requirements were met for aducanumab, with substantial evidence that the drug reduces amyloid beta plaque, and that this reduction is reasonably likely to predict clinical benefit. For drugs approved using the accelerated approval pathway, further study is required to verify anticipated clinical benefits
FDA is preoccupied and determined to approve aducanumab no matter which pathway is used. The following conclusion is subjective and a lot of people will certainly not agree: "We concluded that these requirements were met for aducanumab, with substantial evidence that the drug reduces amyloid-beta plaque, and that this reduction is reasonably likely to predict clinical benefit." Had the FDA been so sure about the biomarker 'amyloid-beta plaque' reduction is 'reasonably likely to predict clinical benefit', they would advise the sponsors (Biogen and other Alzheimer drug developers) to design their phase III studies with the primary efficacy endpoint being the lowering the amyloid-beta plaque, not the measuring the benefit in improving the cognitive and function. 

Accelerated approval pathway is described in FDA guidance for industry "Expedited Programs for Serious Conditions – Drugs and Biologics", but is only used in a situation where the confirmatory studies with clinical endpoints have not been conducted. In Biogen's case, two confirmatory studies with clinical endpoints had already been completed (actually was stopped early for futility). It is a round peg in a square hole to retrospectively going back to the accelerated approval pathway based on the biomarker because of the conflicting and unconvincing results from confirmatory trials with clinical endpoints. Approval of aducanumab based on an accelerated approval pathway breaks agency precedent. "Accelerated approval is traditionally used for treatments that haven't yet proved themselves in large trials. In Biogen's case, Aduhelm went through two Phase 3 studies and came up with conflicting evidence."

FDA also loses its fairness - there are a lot of diseases with unmet medical needs. The drugs for other unmet medical conditional have been tested and generated stronger evidence than Biogen's pivotal studies, but the drugs were rejected by FDA. Here is an article about ALS (amyotrophic lateral sclerosis) - more deadly than Alzheimer's disease.

FDA's controversial Aduhelm decision leaves ALS patients feeling spurned

The FDA's controversial approval of Biogen's Aduhelm drug for Alzheimer's disease has been met with fierce resistance from all corners of the biopharma industry, but few seem to be as upset with the decision as ALS patients and advocacy groups.

For all that's already been written and discussed about the agency's announcement, from the drug's exorbitantly high price of $56,000 per year to criticism over lowered standards, ALS patients see something more. ALS patients and associations say they largely regarded Aduhelm's approval as a bittersweet double standard: happy that those with Alzheimer's have a new drug available, but questioning how the FDA evaluated Biogen's drug compared to the experimental programs being studies for their own disease. 

Nothing punctuated the feeling harder than the agency's announcement in April that a promising drug under development by the biotech Amylyx would need another study to confirm efficacy. This program, called AMX0035, hit the primary endpoint for improving function specifically laid out in the FDA's 2019 guidelines for new ALS treatments, whereas Biogen halted two pivotal Aduhelm studies early because of futility in its own function measurements. 

In general, to demonstrate substantial evidence of effectiveness of the drug, two adequate and well-controlled trials are needed. In Biogen's case, two adequate and well-controlled trials ENGAGE and EMERGE to evaluate the efficacy and safety of aducanumab in patients. When two studies gave contradicting results (one positive and one not positive), a third adequate and well-controlled study will be needed (before the drug approval, not after the drug approval). I remembered other examples: Pirfenidone was developed for treating the rare disease of IPF (idiopathic pulmonary fibrosis). The sponsor conducted two pivotal studies with one study positive (p=0.01) and one study negative (p=0.5). Initial NDA submission with these two studies was rejected by the FDA. FDA demanded the sponsor to conduct a third study. A third study gave a positive result (p<0.01) and NDA was resubmitted, and FDA approved the Perfenidone for IPF. Another example is ciprofloxacin dispersion in non-CF bronchiectasis (rare disease without approved treatment). The sponsor conducted two identical phase III studies ORIBIT-3 and ORBIT-4 - two studies gave contradicting results (one positive and one not positive). The NDA was rejected by FDA and additional studies were not conducted due to funding issues - ciprofloxacin dispersion remains not approved for non-CF bronchiectasis. 

In Biogen's case, two years have passed since they revealed the results of their pre-maturely discontinued studies: one with positive and one with negative. They could have started the third study and would be able to complete the third study not far from now. Instead, with FDA's help, they got their aducanumab approved without doing the third study and they were given a long 9-years to do a post-marketing phase IV study.

References:

Tuesday, June 01, 2021

Decentralized clinical trials and in silico clinical trials

These days, the buzzword in the clinical trial field is 'decentralized clinical trials' or DCTs. The Covid-19 pandemic seems to push the clinical trials toward 'decentralized' or 'hybrid' of decentralized and traditional clinical trials. 

Traditional clinical trials are 'centered' around the clinical trial sites and the investigators. The patients (clinical trial participants) are recruited by the investigators who are the medical doctors responsible for the conduct of the clinical trial at trial sites. The trial sites are the clinics, hospitals, and medical centers. The patients would need to visit the trial sites regularly to see the investigators for clinical trial activities (signing the informed consent, screening for eligibility, receiving study treatments, performing efficacy and safety measures,...). The clinical trial data will then be recorded and entered into the database (for example EDC) by the study coordinator or investigator at investigational sites.

Decentralized clinical trials are defined as the decentralization of clinical trial operations where technology is used to communicate with study participants and collect data and the data collection will not depend on the frequent patient's visits to the investigational sites. According to CTTI (clinical trial transformation initiative) Recommendations: Decentralized Clinical Trials
DCTs using telemedicine and other emerging and novel information technology (IT)
services offer the potential for local HCPs to participate in clinical trials. This may
provide several advantages compared to traditional clinical trials conducted at more
centralized clinical trial sites, including the following:
  • Faster trial participant recruitment, which can accelerate trial participant access to important medical interventions and reduce costs for sponsors.
  • Improved trial participant retention, which may reduce missing data, shorten clinical trial timelines, and improve data interpretability.
  • Greater control, convenience, and comfort for trial participants by offering at home or local patient care.
  • Increased diversity of the population enrolled in clinical trials.
  • An opportunity for home administration or home use of the IMP, which may be
  • more representative of real-world administration/use post-approval.
FDA's Advancing Oncology Decentralized Trials - Learning from COVID-19 Trial Datasets also listed the advantages of the DCTs: 
  • Decentralized Clinical Trials (DCT) may have several potential benefits including reduced patient and sponsor burden and increased accrual and retention of a more diverse trial population.
  • Use of full or hybrid DCT designs by commercial sponsors has been rare in oncology, in part due to uncertainty surrounding the effect of remote assessments on data quality and outcomes.
  • COVID-19 has necessitated DCT-type trial modifications such as remote assessments to reduce patient exposure to COVID-19 infection from travel to trial sites.
  • Many of these remote assessment modifications were deployed in the middle of large ongoing cancer trials.
  • There is an opportunity to evaluate the effect of remote assessments on trial data to advance Decentralized Trials in oncology.
  • Better understanding of the effect of DCT modifications can reduce uncertainty for sponsors and regulatory bodies, and identify mitigation strategies for future prospective DCT designs.
Historically, DCTs may be called differently: virtual trials, siteless trials, remote trials, digital trials, direct-to-patient trials. These different names may just describe one specific aspect of the DCTs and can cause confusion. For example, 'virtual trials' can be confused with the in silico trials which are based on computer models and do not use real participants (patients) but computer programs to model participants to assess drug efficacy and safety during the preclinical phase or before a traditional trial.

"Decentralized clinical trial" is a “Terrible Name for a Promising Innovation” and is not a best terminology for patients and study participants. Alternative names such as direct-to-patient trials, patient-centric trials, and home-based trials seem to be more straightforward and better terms.

In essence, in traditional clinical trials, we bring the trial to the patients; in DCTs, we bring the patients to the trial.

There are still a lot of challenges and obstacles to implementing decentralized clinical trials. Application of DCTs may be limited to some special situations (such as post-marketing studies with patient-reported outcomes and outcomes measured digitally). It is still rare for pivotal and registration studies to use full DCTs. It seems to be more appropriate to adopt hybrid trials - the combination of the traditional and the decentralized trials. For example, in clinical trials in the rare disease area, it is difficult for patients to travel to the investigational sites, the patients may visit the investigational sites for some important visits (in-clinic visits) and then the home health care nurses may be used for in-home visits to patient's home.  


In 2019, Janssen, PRA Launch a Fully Virtual Trial (it should be called the decentralized trial) "A Study on Impact of Canagliflozin on Health Status, Quality of Life, and Functional Status in Heart Failure (CHIEF-HF)". The study design was described in the Circulation: Heart Failure "Novel Trial Design: CHIEF-HF". CHIEF-HF seems to be the first phase III trial being fully decentralized.
 
Here are some references for DCTs:   
The decentralized clinical trials still collect the data from the patients and should not be called 'virtual clinical trials'. On the contrary, the 'In Silico clinical trials' is more appropriately called 'virtual clinical trials' or 'patientless clinical trials' and it simulates the virtual subjects for modeling and prediction. In Silico clinical trials may use the data from pre-clinical and historical clinical trials for simulation but involves no real patients in the study.  

According to the senator bill "AGRICULTURE, RURAL DEVELOPMENT, FOOD AND DRUG
ADMINISTRATION, AND RELATED AGENCIES APPROPRIATIONS BILL, 2016", In Silico clinical trials use computer models and simulations to develop and assess devices and drugs, including their potential risk to the public, before being tested in live clinical trials."
 

In Silico clinical trials are part of the model-informed drug development (MIDD). the FDA has a MIDD pilot program managed by the Division of Pharmacometrics

Dr. Yaning Wang has multiple presentations promoting the MIDD and In Silico clinical trials, for example, in his presentation at the 2021 FDA Science Forum "Regulatory Applications and Research of Model-Informed Drug Development (MIDD)" (Youtube video at 2:38:25) and in his presentation at PMDA "Application of MIDD in New Drug Development and Approval". 

Here are some additional references on In Silico clinical trials: