merged in past version of presentation. Turns out I had left it on git and had not merged it.
Will King
3 years ago
commit
5a27e4a567
@ -1 +1 @@
|
||||
Subproject commit a2c0e4dcc70a70041e4895698c9dd856defdb7ed
|
||||
Subproject commit 05a96a3a29861e682f01498c5499eb686d064409
|
||||
@ -0,0 +1,24 @@
|
||||
# Intro
|
||||
|
||||
# Lit Review
|
||||
|
||||
# Causal Identification
|
||||
https://xkcd.com/2726/
|
||||
|
||||
Because running an experimental trial on how clinical trial recruitment and
|
||||
drugs on market affect clinical trial completion is going to be nigh impossible.
|
||||
Finding natural experiments may also be difficult.
|
||||
Instead going to use a structural approach based on Pearl's Do-Calculus.
|
||||
|
||||
Background on backdoor criterion. #can be ignored in this draft?
|
||||
|
||||
Present Graph
|
||||
|
||||
Discuss adjustment sets for total vs direct effects.
|
||||
|
||||
# Conclusion
|
||||
|
||||
# Appendix
|
||||
Include table of ?? Hierarchy
|
||||
|
||||
# References
|
||||
@ -1,19 +1,97 @@
|
||||
\documentclass[../Main.tex]{subfiles}
|
||||
\graphicspath{{\subfix{Assets/img/}}}
|
||||
|
||||
%%%%% Organization %%%%%
|
||||
% First read and write personal notes.
|
||||
% Then organizing/tagging the literature along these axes
|
||||
% - Trial and Trial Sequence Completion/success rates
|
||||
% - R&D efforts
|
||||
% - FIND MORE
|
||||
% -
|
||||
%
|
||||
% Now figure out the order in which to present them and what they explain.
|
||||
% Now sumarize each one, within the context of the literature around it.
|
||||
% Now write them out together
|
||||
\begin{document}
|
||||
TODO
|
||||
|
||||
This paper sits within an intersection of health and industrial organization economics
|
||||
that is frequently studied.
|
||||
Encouraging a strong supply of novel and generic pharmaceuticals contributes
|
||||
in important ways to both public health and fiscal policy.
|
||||
Not only to the pathway to drug approval long, as many as 90\% of compounds
|
||||
that begin human trials fail to gain approval
|
||||
(\cite{khmelnitskaya_competition_2021}).
|
||||
Complicating this is the complex regulatory and competitive environment in
|
||||
which pharmaceutical companies operate.
|
||||
|
||||
%%%%%%%%% Why are drugs so expensive?
|
||||
|
||||
% van der Grond, Uyle-de Groot, Pieters 2017
|
||||
% - What causes high costs of drugs?
|
||||
% - High level synthesis of discussion regarding causes
|
||||
% - Academic and non-academic sources
|
||||
|
||||
|
||||
%%%%%%%%%%%%%%%% What do we know about clinical trials?
|
||||
|
||||
% Hwang, Carpenter, Lauffenburger, et al (2016)
|
||||
% - Why do investigational new drugs fail during late stage trials?
|
||||
\citeauthor{hwang_failure_2016} (\citeyear{hwang_failure_2016})
|
||||
investigated causes for which late stage (Phase III)
|
||||
clinical trials fail across the USA, Europe, Japan, Canada, and Australia.
|
||||
They found that for late stage trials that did not go on to recieve approval,
|
||||
57\% failed on efficacy grounds, 17\% failed on safety grounds, and 22\% failed
|
||||
on commercial or other grounds.
|
||||
For context, this current work hopes to be able to distinguish some of the
|
||||
mechanisms behind those commercial or other failures.
|
||||
|
||||
% Abrantes-Metz, Adams, Metz (2004)
|
||||
% - What correlates with successfully passing clinical trials and FDA review?
|
||||
% -
|
||||
In \citeyear{abrantes-metz_pharmaceutical_2004},
|
||||
\citeauthor{abrantes-metz_pharmaceutical_2004}
|
||||
described the relationship between
|
||||
various drug characteristics and how the drug progressed through clinical trials.
|
||||
This non-causal estimate was notable for using a
|
||||
mixed state proportional hazard model and estimating the impact of
|
||||
observed characteristics in each of the three phases.
|
||||
They found that as trials last longer, the rate of failure increases for
|
||||
Phase I \& II trials, while Phase 3 trials generally have a higher rate of
|
||||
success than failure after 91 months.
|
||||
|
||||
% Ekaterina Khmelnitskaya (2021)
|
||||
% - separates scientific from market failure of the clinical drug pipeline
|
||||
In her doctoral dissertation, Ekaterina Khmelnitskaya studied the transition of
|
||||
drug candidates between clinical trial phases.
|
||||
Her key contribution was to find ways to disentangle strategic exits from the
|
||||
development pipeline and exits due to clinical failures.
|
||||
She found that overall 8.4\% of all pipeline exits are due to strategic
|
||||
terminations and that the rate of new drug production would be about 23\%
|
||||
higher if those strategic terminatations were elimintated
|
||||
(\cite{khmelnitskaya_competition_2021}).
|
||||
|
||||
% Waring, Arrosmith, Leach, et al (2015)
|
||||
% - Atrition of drug candidates from four major pharma companies
|
||||
% - Looked at how phisicochemical properties affected clinical failure due to safety issues
|
||||
%not in this version
|
||||
|
||||
|
||||
|
||||
|
||||
%%%%%%%%% What do we know about drug development incentives?
|
||||
|
||||
% Dranov, Garthwaite, and Hermosilla (2022)
|
||||
% - does the demand-pull theory of R&D explain novel compound development?
|
||||
% - no, it is biased towards follow-on drug R&D.
|
||||
|
||||
% Acemoglu and Linn
|
||||
% - Market size in innovation
|
||||
% - Exogenous demographic trends has a large impact on the entry of non-generic drugs and new molecular entitites.
|
||||
On the side of market analysis, %TODO:remove when other sections are written up.
|
||||
\citeauthor{acemoglu_market_2004}
|
||||
(\citeyear{acemoglu_market_2004})
|
||||
used exogenous deomographics changes to show that the
|
||||
entry of novel compounds is highly driven by the underlying aged population.
|
||||
They estimate that a 1\% increase in applicable demographics increase the
|
||||
entry of new drugs by 6\%, mostly concentrated among generics.
|
||||
Among non-generics, a 1\% increase in potential market size
|
||||
(as measured by demographic groups) leads to a 4\% increase in novel therapies.
|
||||
|
||||
% Gupta
|
||||
% - Inperfect intellectual property rights in the pharmaceutical industry
|
||||
%\cite{GupaPhd2023}
|
||||
|
||||
% Agarwal and Gaule 2022
|
||||
% - Retrospective on impact from COVID-19 pandemic
|
||||
% Not in this version
|
||||
|
||||
\end{document}
|
||||
|
||||
@ -1,12 +1,115 @@
|
||||
\documentclass[../Main.tex]{subfiles}
|
||||
\graphicspath{{\subfix{Assets/img/}}}
|
||||
|
||||
\begin{document}
|
||||
In September of 2019, the European Space Agency (ESA) released a tweet
|
||||
explaining that they had performed an appendicotomy in space using
|
||||
nothing more than radiation from the sun.
|
||||
They are sad to announce that the patient died due to complications from
|
||||
exposure to the cold vaccum of space.
|
||||
%\subsection{Data Exploration} %TODO: fill this out later.
|
||||
%look at trial
|
||||
\subsection{Model Fitting}
|
||||
In this section we examine the results from fitting the econometric model using
|
||||
mc-stan (\cite{mc-stan}) through the rstan (\cite{rstan}) interface.
|
||||
|
||||
%describe
|
||||
The model was based on the hierarchal logistic regression model
|
||||
presented in the Stan Users Guide (\cite{mc-stan}),
|
||||
and was run with 2,500 warmup iterations and
|
||||
2,500 sampling iterations in six chains.
|
||||
There were various issues, including 160 divergent transitions and the R-hat
|
||||
measure was 1.49.
|
||||
Overall these suggest that the econometric model is incorrect as
|
||||
written or requires reparameterization.
|
||||
%TODO: and info about how I learned about these diagnostics
|
||||
|
||||
|
||||
\subsubsection{Diagnostics}
|
||||
%Examine trank plots
|
||||
To identify which parameters were problematic, I first looked at trace rank
|
||||
histograms.
|
||||
Under idea circumstances, each line (representing a chain) should exchange
|
||||
places with the other lines frequently.
|
||||
In both \cref{fig:mu_trank} and \cref{fig:sigma_trank}, most parameters seem
|
||||
to mix well but there are a couple of exceptions.
|
||||
This warrants further investigation.
|
||||
|
||||
\begin{figure}[H]
|
||||
\includegraphics[width=\textwidth]{../assets/img/mu_trank.png}
|
||||
\caption{Trace Rank Histogram: Mu values}
|
||||
\label{fig:mu_trank}
|
||||
\end{figure}
|
||||
|
||||
\begin{figure}[H]
|
||||
\includegraphics[width=\textwidth]{../assets/img/sigma_trank.png}
|
||||
\caption{Trace Rank Histogram: Sigma values}
|
||||
\label{fig:sigma_trank}
|
||||
\end{figure}
|
||||
|
||||
%Take a look at batman and points for mu
|
||||
In the case of the Mu values, a parallel coordinates plot
|
||||
doesn't seem to indicate any parameters as likely candidates
|
||||
for causing the issues with divergent transitions.
|
||||
\begin{figure}[H]
|
||||
\includegraphics[width=\textwidth]{../assets/img/mu_batman.png}
|
||||
\caption{Parallel Coordinate Plot: Mu values}
|
||||
\label{fig:mu_batman}
|
||||
\end{figure}
|
||||
Note that at each parameter, there is some level of dispersion between
|
||||
values that diverged.
|
||||
|
||||
On the other hand, in the parallel coordinates plot for sigma values,
|
||||
it appears that most divergent transitions occur with values of
|
||||
sigma[1], sigma[3], sigma[6], and sigma[7] close to zero.
|
||||
\begin{figure}[H]
|
||||
\includegraphics[width=\textwidth]{../assets/img/sigma_batman.png}
|
||||
\caption{Parallel Coordinate Plot: Sigma values}
|
||||
\label{fig:sigma_batman}
|
||||
\end{figure}
|
||||
Overall this suggests that there is an issue with the specification
|
||||
of the covariance structures of the hyperparameters.
|
||||
|
||||
Additional evidence that the covariance structure is incorrect comes from
|
||||
plotting pairs of parameter values and examining the chains with divergent
|
||||
transitions.
|
||||
|
||||
\begin{figure}[H]
|
||||
\includegraphics[width=\textwidth]{../assets/img/sigma_pairs_5-9.png}
|
||||
\caption{Parameter Pairs plots: Sigma[5] through Sigma[9]}
|
||||
\label{fig:sigma_pairs_5-9.png}
|
||||
\end{figure}
|
||||
From this we can see that divergent pairs are highly correlated with the cases
|
||||
where sigma[6] or sigma[7] are equal to zero.
|
||||
This has an impact on the shape of both of those estimated parameters, causing
|
||||
both to be bimodal.
|
||||
|
||||
|
||||
\subsection{Interpretation}
|
||||
|
||||
Ignoring the diagnosed issues with the model, we do see some interesting
|
||||
preliminary results.
|
||||
|
||||
%in mu, mu[5] shifted strongly
|
||||
In \cref{fig:mu_posterior} we see that mu[5], the parameter corresponding
|
||||
to enrollment appears to be strongly negative.
|
||||
This is consistent with the idea that enrollment close to planned enrollment
|
||||
decreases the probability of terminating the trial.
|
||||
In \cref{fig:sigma_posterior}, sigma[2] (corresponding to the number of brands
|
||||
selling the drug of interest) has a large variance covers some relatively
|
||||
high values.
|
||||
This suggests that the impact of how frequently the drug is sold varies greatly
|
||||
across different ICD-10 categories of disease.
|
||||
|
||||
|
||||
\begin{figure}[H]
|
||||
\includegraphics[width=\textwidth]{../assets/img/mu_posterior.png}
|
||||
\caption{Posterior Parameter Estimates: Mu}
|
||||
\label{fig:mu_posterior}
|
||||
\end{figure}
|
||||
|
||||
% Sigma[2] suggests there is a high variance in the impact that the number of drugs on the market has.
|
||||
\begin{figure}[H]
|
||||
\includegraphics[width=\textwidth]{../assets/img/sigma_posterior.png}
|
||||
\caption{Posterior Hyperparameter Estimates: Sigma}
|
||||
\label{fig:sigma_posterior}
|
||||
\end{figure}
|
||||
|
||||
Due to the deficiencies in the data and model, this is the limit of the
|
||||
analysis I will perform at this time.
|
||||
|
||||
\end{document}
|
||||
|
||||
@ -0,0 +1,8 @@
|
||||
\documentclass[../Main.tex]{subfiles}
|
||||
\graphicspath{{\subfix{Assets/img/}}}
|
||||
|
||||
\begin{document}
|
||||
\subsection{Appendix 1}\label{Appendix1}
|
||||
Insert a table containing the GBD Data Here
|
||||
|
||||
\end{document}
|
||||
@ -0,0 +1,101 @@
|
||||
\documentclass[../Main.tex]{subfiles}
|
||||
\graphicspath{{\subfix{Assets/img/}}}
|
||||
|
||||
\begin{document}
|
||||
|
||||
As noted above, there are various issues with the analysis as completed so far.
|
||||
Below I discuss various steps that I believe will improve the analysis.
|
||||
|
||||
\subsection{Increasing number of observations}
|
||||
|
||||
The most important step is to increase the number of observations available.
|
||||
Currently this requires matching trials to ICD-10 codes by hand, but
|
||||
there are certainly some steps that can be taken to improve the speed with which
|
||||
this can be done.
|
||||
|
||||
\subsection{Covariance Structure}
|
||||
|
||||
As noted in the diagnostics section, many of the convergence issues seem
|
||||
to occure in the covariance structure.
|
||||
Instead of representing the parameters $\beta$ as independently normal:
|
||||
\begin{align}
|
||||
\beta_k(d) \sim \text{Normal}(\mu_k, \sigma_k)
|
||||
\end{align}
|
||||
I propose using a multivariate normal distribution:
|
||||
\begin{align}
|
||||
\beta(d) \sim \text{MvNormal}(\mu, \Sigma)
|
||||
\end{align}
|
||||
I am not familiar with typical approaches to priors on the covariance matrix,
|
||||
so this will require a further literature search as to best practices.
|
||||
|
||||
\subsection{Finding Reasonable Priors}
|
||||
|
||||
In standard bayesian regression, heavy tailed priors are common.
|
||||
When working with a bayesian bernoulli-logit model, this is not appropriate as
|
||||
heavy tails cause the estimated probabilities $p_n$ to concentrate around the
|
||||
values $0$ and $1$, and away from values such as $\frac{1}{2}$ as discussed in
|
||||
\cite{mcelreath_statistical_2020}. %TODO: double check the chapter for this.
|
||||
|
||||
I indend to take the general approach recommended in \cite{mcelreath_statistical_2020} of using
|
||||
prior predictive checks to evaluate the implications of different priors
|
||||
on the distribution on $p_n$.
|
||||
This would consist of taking the independent variables and predicting the values
|
||||
of $p_n$ based on a proposed set of priors.
|
||||
By plotting these predictions, I can ensure that the specific parameter priors
|
||||
used are consistent with my prior beliefs on how $p_n$ behaves.
|
||||
Currently I believe that $p_n$ should be roughly uniform or unimodal, centered
|
||||
around $p_n = \frac{1}{2}$.
|
||||
|
||||
|
||||
\subsection{Imputing Enrollment}
|
||||
|
||||
Finally, I must address the issue of how enrollment is reported.
|
||||
In many cases, the trial continues to report an anticipated enrollment value
|
||||
while the trial is still recruiting.
|
||||
Thus using anticipated enrollment figures is inappropriate.
|
||||
I am planning on using bayesian imputation to estimate actual enrollment
|
||||
when it has not yet occured.
|
||||
This will require building a statistical model of the enrollment process.
|
||||
One advantage this dataset has is that trial sponsors provide their anticipated
|
||||
enrollment numbers, allowing me to use this in the prediction model.
|
||||
Additionally, each snapshot contains the elapsed duration and current status of
|
||||
the trial , which may help improve the prediction.
|
||||
Although predicted enrollment will be imprecise, it explicitly accounts for
|
||||
uncertanty in the imputation and dependent calculations \cite{mcelreath_statistical_2020}.
|
||||
|
||||
\subsection{Improving Population Estimates}
|
||||
|
||||
The Global Burden of Disease dataset contains the best estimates of disease
|
||||
population sizes that I have found so far.
|
||||
Unfortunately, for some conditions it can be relatively imprecise due to
|
||||
its focus on providing data geared towards public health policy.
|
||||
For example, GBD contains categories for both
|
||||
drug resistant and drug suceptible tuberculosis.
|
||||
In contrast, there is no category for non-age related macular degeneration.
|
||||
One resulting concern is that for a given ICD-10 code, the applicable GBD population
|
||||
estimates may act as an estimate of the upper bound of population size
|
||||
(\cite{global_burden_of_disease_collective_network_global_2020}). %fix citation
|
||||
I would like to explicitly address this in my model, although I have not
|
||||
found a way to do so.
|
||||
|
||||
|
||||
\subsection{Improving Measures of Market Conditions}
|
||||
|
||||
Finally, the currently employed measure of market conditions -- the number of
|
||||
brands using the same active ingredients -- is not a very good measure of
|
||||
the options available to potential participants of a clinical trial.
|
||||
The ideal measures would capture the alternatives available to treat a given
|
||||
disease (drug meeting the given indication) at the time of the trial snapshot,
|
||||
but this data is hard to come by.
|
||||
In addition to the fact that many diseases may be treated by non-pharmaceutical
|
||||
means, off-label prescription of pharmaceuticals is legal at the federal level
|
||||
(\cite{commissioner_understanding_2019}).
|
||||
These two facts both complicate measuring market conditions.
|
||||
|
||||
One dataset that I have only investigated briefly is the \url{DrugCentral.org}
|
||||
database which tracks official indications and some off-label indications as
|
||||
well
|
||||
(\cite{ursu_drugcentral_2017}).
|
||||
|
||||
|
||||
\end{document}
|
||||
@ -0,0 +1,14 @@
|
||||
\documentclass[../Main.tex]{subfiles}
|
||||
\graphicspath{{\subfix{Assets/img/}}}
|
||||
|
||||
\begin{document}
|
||||
Identifying commercial impediments to successfully completing
|
||||
clinical trials in otherwise capable pharmaceuticals will hopefully
|
||||
lead to a more robust and competitive market.
|
||||
Although the current state of this research is insufficient to draw robust
|
||||
conclusions, early results suggest that enrollment rates have some impact
|
||||
on whether or not a clinical trial terminates early or continues
|
||||
to full completion.
|
||||
|
||||
|
||||
\end{document}
|
||||
Binary file not shown.
@ -0,0 +1,573 @@
|
||||
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
||||
% Beamer Presentation
|
||||
% LaTeX Template
|
||||
% Version 1.0 (10/11/12)
|
||||
%
|
||||
% This template has been downloaded from:
|
||||
% http://www.LaTeXTemplates.com
|
||||
%
|
||||
% License:
|
||||
% CC BY-NC-SA 3.0 (http://creativecommons.org/licenses/by-nc-sa/3.0/)
|
||||
%
|
||||
% Changed theme to WSU by William King
|
||||
%
|
||||
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
||||
|
||||
%----------------------------------------------------------------------------------------
|
||||
% PACKAGES AND THEMES
|
||||
%----------------------------------------------------------------------------------------
|
||||
|
||||
\documentclass[xcolor=dvipsnames,aspectratio=169]{beamer}
|
||||
|
||||
|
||||
%Import Preamble bits
|
||||
\input{../assets/preambles/FormattingPreamble.tex}
|
||||
\input{../assets/preambles/TikzitPreamble.tex}
|
||||
\input{../assets/preambles/MathPreamble.tex}
|
||||
\input{../assets/preambles/BibPreamble.tex}
|
||||
\input{../assets/preambles/GeneralPreamble.tex}
|
||||
|
||||
|
||||
|
||||
|
||||
%----------------------------------------------------------------------------------------
|
||||
% TITLE PAGE
|
||||
%----------------------------------------------------------------------------------------
|
||||
|
||||
\title[Clinical Trials]{The Effects of Market Conditions on Recruitment and Completion of Clinical Trials}
|
||||
|
||||
\author{Will King} % Your name
|
||||
\institute[WSU] % Your institution as it will appear on the bottom of every slide, may be shorthand to save space
|
||||
{
|
||||
Washington State University \\ % Your institution for the title page
|
||||
\medskip
|
||||
\textit{william.f.king@wsu.edu} % Your email address
|
||||
}
|
||||
\date{\today} % Date, can be changed to a custom date
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
\begin{document}
|
||||
\begin{frame}
|
||||
\titlepage % Print the title page as the first slide
|
||||
\end{frame}
|
||||
%----------------------------------
|
||||
\begin{frame} %Allow frame breaks
|
||||
\frametitle{Clincial Trials} % Table of contents slide, comment this out to remove it
|
||||
% - Intro and hook (Clinical Trials are key part of pharmacological pipeline)
|
||||
Pharmaceuticals are a frequently discussed aspect of health care cost managment.
|
||||
Their development is dictated by scientific and regulatory hurdles
|
||||
including passing clinical trials
|
||||
(\cite{noauthor_fda_nodate}),
|
||||
while their market is characterized by strategic competition and ambiguous
|
||||
patent protection
|
||||
(\cite{van_der_gronde_addressing_2017}).
|
||||
|
||||
\vspace{12pt}
|
||||
|
||||
This research investigates the pathways by which market conditions
|
||||
affect clinical trial completion.
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{This research}
|
||||
\textbf{Questions:}
|
||||
\begin{enumerate}
|
||||
\item Does the existence of alternative drugs on the market make it
|
||||
harder for clinical trials to complete successfully?
|
||||
\item How much of this is occurs due to increased recruitment difficulty?
|
||||
\end{enumerate}
|
||||
|
||||
\end{frame}
|
||||
%--------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Thanks} % Table of contents slide, comment this out to remove it
|
||||
Thanks to Chris Adams and Rebecca Sachs of the Congressional Budget Office.
|
||||
\end{frame}
|
||||
%--------------------------------
|
||||
\begin{frame}[allowframebreaks] %Allow frame breaks
|
||||
\frametitle{Overview} % Table of contents slide, comment this out to remove it
|
||||
\tableofcontents
|
||||
% - Intro and hook
|
||||
% - Literature review
|
||||
% - Causal Identification
|
||||
% - Data
|
||||
% - Econometric model
|
||||
% - Results
|
||||
% - Improvements
|
||||
\end{frame}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Lit Review %%%%%%%%%%%%%%%%%%%%%%%%
|
||||
\section{Lit Review}
|
||||
% First slide:
|
||||
%-------------------------------------------------------------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Literature Highlights}
|
||||
\begin{itemize}
|
||||
\item \cite{van_der_gronde_addressing_2017}:
|
||||
High level synthesis of overall discussion regarding drug costs.
|
||||
Both academic and non-academic sources.
|
||||
\item \cite{hwang_failure_2016}:
|
||||
Answered the question "Why do late-stage (phase III) trials fail?"
|
||||
Found that efficacy, safety, and competition reasons accounted for
|
||||
57\%, 17\%, and 22\% respectively.
|
||||
\item \cite{abrantes-metz_pharmaceutical_2004}:
|
||||
Described how drugs progress through the 3 phases of clinical trials
|
||||
and correllations between various trial characteristics and the
|
||||
clinical trial failures.
|
||||
\item \cite{khmelnitskaya_competition_2021}:
|
||||
Modeled clinical trial lifecycle of drugs, found method to separate
|
||||
scientific from competitive reasons for failure to progress to the
|
||||
next phase.
|
||||
% \item \cite{}:
|
||||
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{This research, in context}
|
||||
|
||||
In contrast to previous work looking at multiple phases of trials,
|
||||
I seek to figure out what causes individual trials to fail.
|
||||
|
||||
\vspace{12pt}
|
||||
|
||||
Instead of focusing on the drug development pipeline, I attempt to
|
||||
investigate the population of drug-based, phase III trials.
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame} %Allow frame breaks
|
||||
\frametitle{Why this approach?} % Table of contents slide, comment this out to remove it
|
||||
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.8\textheight]{../assets/img/methodology_trial.png}
|
||||
\label{FIG:xkcd2726}
|
||||
\caption{``If you think THAT'S unethical, you should see the stuff we approved via our Placebo IRB.''
|
||||
- \url{https://xkcd.com/2726}
|
||||
}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Causal Identification / DGP%%%%%%%%%%%%%%%%%%%%%%%%
|
||||
\section{Causal Model}
|
||||
% Data Generating process
|
||||
% - Agents and their decisions
|
||||
% - Factors that influence each decision
|
||||
% -
|
||||
% -
|
||||
%-------------------------------------------------------------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Data Generating Process}
|
||||
% study sponsors
|
||||
Study Sponsors Decide to start a Phase 3 trial and whether to terminate it.
|
||||
\\
|
||||
They ask themselves:
|
||||
\begin{itemize}
|
||||
\item Do safety incidents require terminating a trial?
|
||||
\item Do efficacy results indicate the trial is worth continuing?
|
||||
\item Is recruiting sufficient to achieve our results and contain costs?
|
||||
\item Do expectations about future returns justify our expenditures?
|
||||
\end{itemize}
|
||||
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Data Generating Process}
|
||||
% participants
|
||||
Participants decide to enroll (and disenroll) themselves in a trial based
|
||||
\begin{itemize}
|
||||
\item Disease severity
|
||||
\item Relative safety/efficacy compared to other treatments
|
||||
\end{itemize}
|
||||
|
||||
Study sponsors plan their enrollment considering
|
||||
\begin{itemize}
|
||||
\item Total population affected
|
||||
\item Likely participant response rates
|
||||
\end{itemize}
|
||||
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Data Generating Process}
|
||||
% Trial Snapshots and dependencies.
|
||||
During a trial, the study sponsor reports snapshots of their trial.
|
||||
This includes updates to:
|
||||
|
||||
\begin{itemize}
|
||||
\item enrollment (actual or anticipated)
|
||||
\item current recruitment status (Recruiting, Active not recruiting, etc)
|
||||
\item study sponsor
|
||||
\item planned completion dates
|
||||
\item elapsed duration
|
||||
\end{itemize}
|
||||
|
||||
Note that final enrollment and the final status (Completed or Terminated)
|
||||
of the trial are jointly determined.
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Causal Diagram: Key Pathways}
|
||||
% Estimating Direct vs Total Effects
|
||||
\begin{figure}
|
||||
\resizebox{!}{0.5\textheight}{
|
||||
\tikzfig{../assets/tikzit/CausalGraph}
|
||||
}
|
||||
\label{FIG:CausalDiagram}
|
||||
\caption{Causal Diagram highlighting direct and total pathways}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Causal Diagram: Backdoor Crieterion}
|
||||
\small
|
||||
\begin{block}{$d$-separation}
|
||||
A set $S$ of nodes blocks a path $p$ if either
|
||||
\begin{enumerate}
|
||||
\item $p$ contains at least one arrow-emitting node in $S$
|
||||
\item $p$ contains at least one collision node $c$ that is outside $S$
|
||||
and has no descendants in $S$.
|
||||
\end{enumerate}
|
||||
If $S$ blocks all paths from X to Y, then it is said to ``$d$-separate''
|
||||
$X$ and $Y$, and then $X \perp Y | S$.
|
||||
\end{block}
|
||||
\begin{block}{Back-Door Criterion}
|
||||
A set $S$ of covariates is admisible as controls on the
|
||||
causal relationship $X \rightarrow Y$ if:
|
||||
\begin{enumerate}
|
||||
\item No element of $S$ is a decendant of $X$
|
||||
\item The elements of $S$ d-separate all paths from $X$ to $Y$ that include
|
||||
parents of $X$.
|
||||
\end{enumerate}
|
||||
\end{block}
|
||||
\cite{pearl_causality_2000}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Causal Diagram}
|
||||
Key takeaways
|
||||
\begin{itemize}
|
||||
\item Measuring enrollment prior to trial completion is necessary for causal identification.
|
||||
\item The backdoor criterion gives us the following adjustment sets:
|
||||
\begin{itemize}
|
||||
\item Total Effect for Market on Termination; Population, Condition, Phase III
|
||||
\item Direct Effects for Enrollment, Market on Termination; Population, Condition Phase III,
|
||||
Elapsed Duration, Planned Enrollment
|
||||
\end{itemize}
|
||||
\item Enrollment requires imputation
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Data %%%%%%%%%%%%%%%%%%%%%%%%
|
||||
\section{Data}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%----------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Sources
|
||||
\subsection{Sources}
|
||||
%----------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame} %Allow frame breaks
|
||||
\frametitle{Data Sources}
|
||||
\begin{itemize}
|
||||
\item ClinicalTrials.gov - AACT \& custom scripts
|
||||
\begin{itemize}
|
||||
\item Select trials of interest
|
||||
\item Trial details:
|
||||
\begin{itemize}
|
||||
\item conditions
|
||||
\item final status
|
||||
\item drugs/interventions
|
||||
\end{itemize}
|
||||
\item Trial snapshots:
|
||||
\begin{itemize}
|
||||
\item enrollment (anticipated, planned, or actual)
|
||||
\item elapsed duration
|
||||
\item current status
|
||||
\end{itemize}
|
||||
\end{itemize}
|
||||
\item Medical Subject Headings (MeSH) Thesaurus
|
||||
\begin{itemize}
|
||||
\item A standardized nomenclature used to classify interventions
|
||||
and conditions in the clinical trials database.
|
||||
\end{itemize}
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame} %Allow frame breaks
|
||||
\frametitle{Data Sources}
|
||||
\begin{itemize}
|
||||
\item NSDE Files (New drug code Structured product labels Data Element)
|
||||
\begin{itemize}
|
||||
\item Contains information about when a given drug was on the market.
|
||||
\end{itemize}
|
||||
\item RxNorm
|
||||
\begin{itemize}
|
||||
\item Links pharmaceuticals between MeSH standardized terms and
|
||||
NSDE files.
|
||||
\end{itemize}
|
||||
\item Global Disease Burden Survey (2019)
|
||||
\begin{itemize}
|
||||
\item Estimates of DALYs for categories of disease
|
||||
\item Links of Categories to ICD-10 Codes
|
||||
\end{itemize}
|
||||
\item ICD-10 (2019)
|
||||
\begin{itemize}
|
||||
\item WHO version
|
||||
\item CMS version (Clinical Managment)
|
||||
\item Used to group disease conditions in hierarchal model
|
||||
\end{itemize}
|
||||
\item Unified Medical Language System Thesaurus
|
||||
\begin{itemize}
|
||||
\item Used to link MeSH standardized terms and ICD-10 conditions
|
||||
\item Manual matching process
|
||||
\end{itemize}
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
%----------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Integration
|
||||
\subsection{Integration}
|
||||
%----------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Data Summaries}
|
||||
%put summaries now
|
||||
\begin{itemize}
|
||||
\item Number of Phase III, FDA monitored Drug Trials: 1,981
|
||||
\item Number of Trials matched to ICD-10: 186
|
||||
\item Number of Trials matched to ICD-10 with population measures: 67
|
||||
(51 completed, 16 terminated)
|
||||
\item Number of Snapshots: 616
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Data used}
|
||||
The following data points were used.
|
||||
\begin{itemize}
|
||||
\item elapsed duration
|
||||
\item asinh(number of brands)
|
||||
\item asinh(high sdi DALY estimate)
|
||||
\item asinh(high-medium sdi DALY estimate)
|
||||
\item asinh(medium sdi DALY estimate)
|
||||
\item asinh(low-medium sdi DALY estimate)
|
||||
\item asinh(low sdi DALY estimate)
|
||||
\end{itemize}
|
||||
The asinh operator was used because it parallells $\text{ln}(x)$ for
|
||||
large values of $x$ but also handles $\text{asinh}(0)=0$.
|
||||
\end{frame}
|
||||
%----------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Summaries: Trial Durations}
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.8\textheight]{../assets/img/2023-04-12_durations_hist.png}
|
||||
\label{FIG:durations}
|
||||
\caption{Trial Durations (days)}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%----------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Summaries: snapshots}
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.8\textheight]{../assets/img/2023-04-12_snapshots_hist.png}
|
||||
\label{FIG:snapshots}
|
||||
\caption{Number of Snapshots per matched trial}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%----------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Summaries: snapshots}
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.8\textheight]{../assets/img/2023-04-12_status_duration_snapshots_points.png}
|
||||
\label{FIG:snapshot_duration_scatter}
|
||||
\caption{Scatterplot of snapshot count and durations}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Econometric Model %%%%%%%%%%%%%%%%%%%%%%%%
|
||||
\section{Econometric model}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Econometric Model}
|
||||
Estimating the total effect of brands on market
|
||||
\begin{align}
|
||||
y_n &\sim \text{Bernoulli}(p_n) \\
|
||||
p_n &= \text{logisticfn}(x_n * \beta(d_n)) \\
|
||||
\beta_k(d) &\sim \text{Normal}(\mu_k, \sigma_k) \\
|
||||
\mu_k &\sim \text{Normal}(0,1) \\
|
||||
\sigma_k &\sim \text{Gamma}(2,1)
|
||||
\end{align}
|
||||
$k$ indexes parameters and $d_n$ represets the ICD-10 group the trial corresponds to.
|
||||
\end{frame}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Results %%%%%%%%%%%%%%%%%%%%%%%%
|
||||
\section{Results}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Results}
|
||||
Because bayesian estimation is typically done numerically, we will first
|
||||
validate convergence.
|
||||
|
||||
Then we will take a look at preliminary results.
|
||||
|
||||
Sampling details
|
||||
\begin{itemize}
|
||||
\item 6 chains
|
||||
\item 2,500 warmup, 2,500 sampling runs
|
||||
\item seed = 11021585
|
||||
\end{itemize}
|
||||
\end{frame}
|
||||
%----------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Convergence Tests
|
||||
\subsection{Convergence}
|
||||
%----------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Warnings}
|
||||
|
||||
\begin{itemize}
|
||||
\item There were no diverging transitions.
|
||||
\item There were 15,000 transitions that exceeded max treedepth.
|
||||
Sampling efficiency is poor.
|
||||
\item All chains had low Bayesian Fraction of Missing Information.
|
||||
Some areas of the distribution were poorly explored.
|
||||
\item R-hat = $1.23$, ideal is around 1, chains did not mix well.
|
||||
\item Bulk and Tail Effective Sample sizes were low,
|
||||
suggesting mean and variance/quantile estimates will be unreliable.
|
||||
\end{itemize}
|
||||
\cite{mc-stan}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Convergence: Mu}
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.9\textheight]{../assets/img/2023-04-11_mu_points.png}
|
||||
\label{FIG:caption}
|
||||
\caption{Hyperparameter Points Plots: Mu}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Convergence: Sigma}
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.8\textheight]{../assets/img/2023-04-11_sigma_points.png}
|
||||
\label{FIG:caption}
|
||||
\caption{Hyperparameter Points Plots: Sigma}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%----------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Preliminary Results
|
||||
\subsection{Preliminary Results}
|
||||
%----------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Preliminary Results: Mu}
|
||||
|
||||
\begin{columns}
|
||||
\begin{column}{0.3\textwidth}
|
||||
\begin{enumerate}
|
||||
\item elapsed duration
|
||||
\item asinh(n\_brands)
|
||||
\item asinh(high sdi)
|
||||
\item asinh(high-medium sdi)
|
||||
\item asinh(medium sdi)
|
||||
\item asinh(low-medium sdi)
|
||||
\item asinh(low sdi)
|
||||
\end{enumerate}
|
||||
\end{column}
|
||||
\begin{column}{0.7\textwidth}
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.8\textheight]{../assets/img/2023-04-11_mu_dist.png}
|
||||
\label{FIG:caption}
|
||||
\caption{Hyperparameter Distribution: Mu}
|
||||
\end{figure}
|
||||
\end{column}
|
||||
\end{columns}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Preliminary Results: Sigma}
|
||||
|
||||
\begin{figure}
|
||||
\includegraphics[height=0.8\textheight]{../assets/img/2023-04-11_sigma_dist.png}
|
||||
\label{FIG:caption}
|
||||
\caption{Hyperparameter Distribution: Sigma}
|
||||
\end{figure}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Interpretation}
|
||||
All of the following interpretations are done in the context of insufficient data
|
||||
|
||||
\begin{enumerate}
|
||||
\item Elapsed Duration (Mu[1]): Trending Negative, reduced probability of termination.
|
||||
\item Number of Brands(Mu[2]): Trending Positive, increased probability of termination.
|
||||
\item Population Measures (Mu[3]-Mu[7])
|
||||
\begin{enumerate}
|
||||
\item What is most surprising is that these are both positive and negative.
|
||||
Probably need more data.
|
||||
\end{enumerate}
|
||||
\item It is surprising to see the wide distribution in sigma values.
|
||||
\end{enumerate}
|
||||
\end{frame}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%%%%%%%%%%%%%%%%%%%% Improvements %%%%%%%%%%%%%%%%%%%%%%%%
|
||||
\section{Improvements}
|
||||
%-------------------------------------------------------------------------------------
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Proposed improvements}
|
||||
\begin{enumerate}
|
||||
\item Match more trials to ICD-10 codes
|
||||
\item Improve Measures of Market Conditions
|
||||
\item Adjust Covariance Structure
|
||||
\item Find Reasonable Priors
|
||||
\item Remove disease categories that don't exist in the data from the priors
|
||||
\item Imputing Enrollment
|
||||
\item Improve Population Estimates
|
||||
\end{enumerate}
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}
|
||||
\frametitle{Questions?}
|
||||
\center{\huge{Questions?}}
|
||||
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\begin{frame}[allowframebreaks]
|
||||
\frametitle{Bibliography}
|
||||
\printbibliography
|
||||
\end{frame}
|
||||
%-------------------------------
|
||||
\end{document}
|
||||
%=========================================
|
||||
%\begin{frame}
|
||||
% \frametitle{MarginalRevenue}
|
||||
% \begin{figure}
|
||||
% \tikzfig{../Assets/owned/ch8_MarginalRevenue}
|
||||
% \includegraphics[height=\textheight]{../Assets/copyrighted/KrugmanObsterfeldMeliz_fig8-7.jpg}
|
||||
% \label{FIG:costs}
|
||||
% \caption{Average Cost Curve as firms enter.}
|
||||
% \end{figure}
|
||||
%\end{frame}
|
||||
%-------------------------------
|
||||
%\begin{frame}
|
||||
% \frametitle{Columns}
|
||||
% \begin{columns}
|
||||
% \begin{column}{0.5\textwidth}
|
||||
% \end{column}
|
||||
% \begin{column}{0.5\textwidth}
|
||||
% \begin{figure}
|
||||
% \tikzfig{../Assets/owned/ch7_EstablishedAdvantageExample2}
|
||||
% \label{FIG:costs}
|
||||
% \caption{Setting the Stage}
|
||||
% \end{figure}
|
||||
% \end{column}
|
||||
% \end{columns}
|
||||
%\end{frame}
|
||||
% %---------------------------------------------------------------
|
||||
Binary file not shown.
|
After Width: | Height: | Size: 156 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 164 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 26 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 27 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 162 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 150 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 164 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 31 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 21 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 64 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 35 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 22 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 65 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 19 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 17 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 24 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 47 KiB |
File diff suppressed because one or more lines are too long
|
After Width: | Height: | Size: 14 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 28 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 200 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 123 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 33 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 86 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 147 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 133 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 149 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 36 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 77 KiB |
File diff suppressed because one or more lines are too long
@ -0,0 +1,30 @@
|
||||
|
||||
@article{van_der_gronde_addressing_2017,
|
||||
title = {Addressing the challenge of high-priced prescription drugs in the era of precision medicine: A systematic review of drug life cycles, therapeutic drug markets and regulatory frameworks},
|
||||
volume = {12},
|
||||
issn = {1932-6203},
|
||||
url = {https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5559086/},
|
||||
doi = {10.1371/journal.pone.0182613},
|
||||
shorttitle = {Addressing the challenge of high-priced prescription drugs in the era of precision medicine},
|
||||
abstract = {Context
|
||||
Recent public outcry has highlighted the rising cost of prescription drugs worldwide, which in several disease areas outpaces other health care expenditures and results in a suboptimal global availability of essential medicines.
|
||||
|
||||
Method
|
||||
A systematic review of Pubmed, the Financial Times, the New York Times, the Wall Street Journal and the Guardian was performed to identify articles related to the pricing of medicines.
|
||||
|
||||
Findings
|
||||
Changes in drug life cycles have dramatically affected patent medicine markets, which have long been considered a self-evident and self-sustainable source of income for highly profitable drug companies. Market failure in combination with high merger and acquisition activity in the sector have allowed price increases for even off-patent drugs. With market interventions and the introduction of {QALY} measures in health care, governments have tried to influence drug prices, but often encounter unintended consequences. Patent reform legislation, reference pricing, outcome-based pricing and incentivizing physicians and pharmacists to prescribe low-cost drugs are among the most promising short-term policy options. Due to the lack of systematic research on the effectiveness of policy measures, an increasing number of ad hoc decisions have been made with counterproductive effects on the availability of essential drugs. Future challenges demand new policies, for which recommendations are offered.
|
||||
|
||||
Conclusion
|
||||
A fertile ground for high-priced drugs has been created by changes in drug life-cycle dynamics, the unintended effects of patent legislation, government policy measures and orphan drug programs. There is an urgent need for regulatory reform to curtail prices and safeguard equitable access to innovative medicines.},
|
||||
pages = {e0182613},
|
||||
number = {8},
|
||||
journaltitle = {{PLoS} {ONE}},
|
||||
shortjournal = {{PLoS} One},
|
||||
author = {van der Gronde, Toon and Uyl-de Groot, Carin A. and Pieters, Toine},
|
||||
urldate = {2023-03-11},
|
||||
date = {2017-08-16},
|
||||
pmid = {28813502},
|
||||
pmcid = {PMC5559086},
|
||||
file = {PubMed Central Full Text PDF:/home/dad/Nextcloud/Zotero_data/storage/7Y8KZSMU/van der Gronde et al. - 2017 - Addressing the challenge of high-priced prescripti.pdf:application/pdf},
|
||||
}
|
||||
@ -0,0 +1,20 @@
|
||||
|
||||
@article{hwang_failure_2016,
|
||||
title = {Failure of Investigational Drugs in Late-Stage Clinical Development and Publication of Trial Results},
|
||||
volume = {176},
|
||||
issn = {2168-6106},
|
||||
url = {http://archinte.jamanetwork.com/article.aspx?doi=10.1001/jamainternmed.2016.6008},
|
||||
doi = {10.1001/jamainternmed.2016.6008},
|
||||
abstract = {{OBJECTIVE} To assess factors associated with regulatory approval or reasons for failure of investigational therapeutics in phase 3 or pivotal trials and rates of publication of trial results. {DESIGN}, {SETTING}, {AND} {PARTICIPANTS} Using public sources and commercial databases, we identified investigational therapeutics that entered pivotal trials between 1998 and 2008, with follow-up through 2015. Agents were classified by therapeutic area, orphan designation status, fast track designation, novelty of biological pathway, company size, and as a pharmacologic or biologic product. {MAIN} {OUTCOMES} {AND} {MEASURES} For each product, we identified reasons for failure (efficacy, safety, commercial) and assessed the rates of publication of trial results. We used multivariable logistic regression models to evaluate factors associated with regulatory approval.
|
||||
{RESULTS} Among 640 novel therapeutics, 344 (54\%) failed in clinical development, 230 (36\%) were approved by the {US} Food and Drug Administration ({FDA}), and 66 (10\%) were approved in other countries but not by the {FDA}. Most products failed due to inadequate efficacy (n = 195; 57\%), while 59 (17\%) failed because of safety concerns and 74 (22\%) failed due to commercial reasons. The pivotal trial results were published in peer-reviewed journals for 138 of the 344 (40\%) failed agents. Of 74 trials for agents that failed for commercial reasons, only 6 (8.1\%) were published. In analyses adjusted for therapeutic area, agent type, firm size, orphan designation, fast-track status, trial year, and novelty of biological pathway, orphan-designated drugs were significantly more likely than nonorphan drugs to be approved (46\% vs 34\%; adjusted odds ratio [{aOR}], 2.3; 95\% {CI}, 1.4-3.7). Cancer drugs (27\% vs 39\%; {aOR}, 0.5; 95\% {CI}, 0.3-0.9) and agents sponsored by small and medium-size companies (28\% vs 42\%; {aOR}, 0.4; 95\% {CI}, 0.3-0.7) were significantly less likely to be approved.
|
||||
{CONCLUSIONS} {AND} {RELEVANCE} Roughly half of investigational drugs entering late-stage clinical development fail during or after pivotal clinical trials, primarily because of concerns about safety, efficacy, or both. Results for the majority of studies of investigational drugs that fail are not published in peer-reviewed journals.},
|
||||
pages = {1826},
|
||||
number = {12},
|
||||
journaltitle = {{JAMA} Internal Medicine},
|
||||
shortjournal = {{JAMA} Intern Med},
|
||||
author = {Hwang, Thomas J. and Carpenter, Daniel and Lauffenburger, Julie C. and Wang, Bo and Franklin, Jessica M. and Kesselheim, Aaron S.},
|
||||
urldate = {2023-01-31},
|
||||
date = {2016-12-01},
|
||||
langid = {english},
|
||||
file = {Hwang et al. - 2016 - Failure of Investigational Drugs in Late-Stage Cli.pdf:/home/dad/Nextcloud/Zotero_data/storage/JJC96CPC/Hwang et al. - 2016 - Failure of Investigational Drugs in Late-Stage Cli.pdf:application/pdf},
|
||||
}
|
||||
@ -0,0 +1,15 @@
|
||||
|
||||
@article{abrantes-metz_pharmaceutical_2004,
|
||||
title = {Pharmaceutical Development Phases: A Duration Analysis},
|
||||
issn = {1556-5068},
|
||||
url = {http://www.ssrn.com/abstract=607941},
|
||||
doi = {10.2139/ssrn.607941},
|
||||
shorttitle = {Pharmaceutical Development Phases},
|
||||
journaltitle = {{SSRN} Electronic Journal},
|
||||
shortjournal = {{SSRN} Journal},
|
||||
author = {Abrantes-Metz, Rosa M. and Adams, Christopher and Metz, Albert D.},
|
||||
urldate = {2023-01-31},
|
||||
date = {2004},
|
||||
langid = {english},
|
||||
file = {Abrantes-Metz et al. - 2004 - Pharmaceutical Development Phases A Duration Anal.pdf:/home/dad/Nextcloud/Zotero_data/storage/LANZBC53/Abrantes-Metz et al. - 2004 - Pharmaceutical Development Phases A Duration Anal.pdf:application/pdf},
|
||||
}
|
||||
@ -0,0 +1,9 @@
|
||||
|
||||
@article{acemoglu_market_2004,
|
||||
title = {{MARKET} {SIZE} {IN} {INNOVATION}: {THEORY} {AND} {EVIDENCE} {FROM} {THE} {PHARMACEUTICAL} {INDUSTRY}},
|
||||
journaltitle = {{QUARTERLY} {JOURNAL} {OF} {ECONOMICS}},
|
||||
author = {Acemoglu, Daron and Linn, Joshua},
|
||||
date = {2004-08},
|
||||
langid = {english},
|
||||
file = {Acemoglu and Linn - MARKET SIZE IN INNOVATION THEORY AND EVIDENCE FRO.pdf:/home/dad/Nextcloud/Zotero_data/storage/HYTY3E36/Acemoglu and Linn - MARKET SIZE IN INNOVATION THEORY AND EVIDENCE FRO.pdf:application/pdf},
|
||||
}
|
||||
@ -0,0 +1,88 @@
|
||||
|
||||
@online{noauthor_fdaaa_nodate,
|
||||
title = {{FDAAA} 801 and the Final Rule - {ClinicalTrials}.gov},
|
||||
url = {https://clinicaltrials.gov/ct2/manage-recs/fdaaa},
|
||||
urldate = {2023-04-08},
|
||||
langid = {english},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/V9YVGVK2/fdaaa.html:text/html},
|
||||
}
|
||||
|
||||
@online{noauthor_frequently_nodate,
|
||||
title = {Frequently Asked Questions - {ClinicalTrials}.gov},
|
||||
url = {https://clinicaltrials.gov/ct2/manage-recs/faq#board},
|
||||
urldate = {2023-04-08},
|
||||
langid = {english},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/GNBZDX5B/faq.html:text/html},
|
||||
}
|
||||
|
||||
@online{noauthor_rxnorm_nodate,
|
||||
title = {{RxNorm}},
|
||||
rights = {Public Domain},
|
||||
url = {https://www.nlm.nih.gov/research/umls/rxnorm/index.html},
|
||||
type = {Product, Program, and Project Descriptions},
|
||||
urldate = {2023-04-08},
|
||||
note = {Publisher: U.S. National Library of Medicine},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/UPPXYYW6/index.html:text/html},
|
||||
}
|
||||
|
||||
@online{noauthor_rxnorm_nodate-1,
|
||||
title = {{RxNorm} Overview},
|
||||
rights = {Public Domain},
|
||||
url = {https://www.nlm.nih.gov/research/umls/rxnorm/overview.html},
|
||||
type = {Product, Program, and Project Descriptions},
|
||||
urldate = {2023-04-08},
|
||||
note = {Publisher: U.S. National Library of Medicine},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/XI269ZNM/overview.html:text/html},
|
||||
}
|
||||
|
||||
@online{noauthor_medical_nodate,
|
||||
title = {Medical Subject Headings - Home Page},
|
||||
rights = {Public Domain},
|
||||
url = {https://www.nlm.nih.gov/mesh/meshhome.html},
|
||||
type = {Product, Program, and Project Descriptions},
|
||||
urldate = {2023-04-09},
|
||||
note = {Publisher: U.S. National Library of Medicine},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/RTW5EPBG/meshhome.html:text/html},
|
||||
}
|
||||
|
||||
@online{noauthor_international_nodate,
|
||||
title = {International Classification of Diseases ({ICD})},
|
||||
url = {https://www.who.int/standards/classifications/classification-of-diseases},
|
||||
abstract = {International Classification of Diseases ({ICD}) Revision},
|
||||
urldate = {2023-04-09},
|
||||
langid = {english},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/4Y3F35AR/classification-of-diseases.html:text/html},
|
||||
}
|
||||
|
||||
@online{noauthor_2023_nodate,
|
||||
title = {2023 {ICD}-10-{CM} {\textbar} {CMS}},
|
||||
url = {https://www.cms.gov/medicare/icd-10/2023-icd-10-cm},
|
||||
urldate = {2023-04-09},
|
||||
}
|
||||
|
||||
@online{noauthor_2023_nodate-1,
|
||||
title = {2023 {ICD}-10-{PCS} {\textbar} {CMS}},
|
||||
url = {https://www.cms.gov/medicare/icd-10/2023-icd-10-pcs},
|
||||
urldate = {2023-04-09},
|
||||
file = {2023 ICD-10-PCS | CMS:/home/dad/Nextcloud/Zotero_data/storage/4NLQJQT6/2023-icd-10-pcs.html:text/html},
|
||||
}
|
||||
|
||||
@online{noauthor_2019_nodate,
|
||||
title = {2019 {ICD}-10-{CM} {\textbar} {CMS}},
|
||||
url = {https://www.cms.gov/Medicare/Coding/ICD10/2019-ICD-10-CM},
|
||||
urldate = {2023-04-09},
|
||||
file = {2019 ICD-10-CM | CMS:/home/dad/Nextcloud/Zotero_data/storage/S5ISTWEL/2019-ICD-10-CM.html:text/html},
|
||||
}
|
||||
|
||||
@online{commissioner_understanding_2019,
|
||||
title = {Understanding Unapproved Use of Approved Drugs "Off Label"},
|
||||
url = {https://www.fda.gov/patients/learn-about-expanded-access-and-other-treatment-options/understanding-unapproved-use-approved-drugs-label},
|
||||
abstract = {Understanding Unapproved Use of Approved Drugs "Off Label"},
|
||||
titleaddon = {{FDA}},
|
||||
author = {Commissioner, Office of the},
|
||||
urldate = {2023-04-10},
|
||||
date = {2019-04-18},
|
||||
langid = {english},
|
||||
note = {Publisher: {FDA}},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/VAKSGTAP/understanding-unapproved-use-approved-drugs-label.html:text/html},
|
||||
}
|
||||
@ -0,0 +1,5 @@
|
||||
|
||||
@misc{noauthor_indexing_nodate,
|
||||
title = {Indexing Spl Fact Sheet},
|
||||
file = {Indexing-SPL-Fact-Sheet.pdf:/home/dad/Nextcloud/Zotero_data/storage/KAHW2ABD/Indexing-SPL-Fact-Sheet.pdf:application/pdf},
|
||||
}
|
||||
File diff suppressed because one or more lines are too long
@ -0,0 +1,7 @@
|
||||
|
||||
@online{noauthor_rxnav---box_nodate,
|
||||
title = {{RxNav}-in-a-Box - {RxNav} Applications},
|
||||
url = {https://lhncbc.nlm.nih.gov/RxNav/applications/RxNav-in-a-Box.html},
|
||||
urldate = {2023-04-10},
|
||||
file = {RxNav-in-a-Box - RxNav Applications:/home/dad/Nextcloud/Zotero_data/storage/A9S2NM29/RxNav-in-a-Box.html:text/html},
|
||||
}
|
||||
@ -0,0 +1,7 @@
|
||||
|
||||
@online{noauthor_icd-10_nodate,
|
||||
title = {{ICD}-10 Version:2019},
|
||||
url = {https://icd.who.int/browse10/2019/en#/C00},
|
||||
urldate = {2023-04-10},
|
||||
file = {ICD-10 Version\:2019:/home/dad/Nextcloud/Zotero_data/storage/23DGMZ5X/en.html:text/html},
|
||||
}
|
||||
@ -0,0 +1,15 @@
|
||||
@Misc{rstan,
|
||||
title = {{RStan}: the {R} interface to {Stan}},
|
||||
author = {{Stan Development Team}},
|
||||
note = {R package version 2.21.8},
|
||||
year = {2023},
|
||||
url = {https://mc-stan.org/},
|
||||
}
|
||||
@Misc{mc-stan,
|
||||
title = {Stan Modelling usersGuide and Reference Manual},
|
||||
author = {{Stan Development Team}},
|
||||
note = {R package version 2.26},
|
||||
year = {2022},
|
||||
url = {https://mc-stan.org/},
|
||||
}
|
||||
|
||||
@ -0,0 +1,13 @@
|
||||
|
||||
@book{mcelreath_statistical_2020,
|
||||
location = {Boca Raton},
|
||||
edition = {2},
|
||||
title = {Statistical rethinking: a Bayesian course with examples in R and Stan},
|
||||
isbn = {978-0-367-13991-9},
|
||||
series = {{CRC} texts in statistical science},
|
||||
shorttitle = {Statistical rethinking},
|
||||
abstract = {"Statistical Rethinking: A Bayesian Course with Examples in R and Stan, Second Edition builds knowledge/confidence in statistical modeling. Pushes readers to perform step-by-step calculations (usually automated.) Unique, computational approach ensures readers understand details to make reasonable choices and interpretations in their modeling work"--},
|
||||
publisher = {Taylor and Francis, {CRC} Press},
|
||||
author = {{McElreath}, Richard},
|
||||
date = {2020},
|
||||
}
|
||||
@ -0,0 +1,9 @@
|
||||
|
||||
@report{global_burden_of_disease_collaborative_network_global_2020-1,
|
||||
location = {Seattle, United States of America},
|
||||
title = {Global Burden of Disease Study 2019 ({GBD} 2019) Cause Hierarchy},
|
||||
abstract = {Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2019 ({GBD} 2019) Cause Hierarchy Seattle, United States of America: Institute for Health Metrics and Evaluation ({IHME}), 2020.},
|
||||
institution = {nstitute for Health Metrics and Evaluation ({IHME})},
|
||||
author = {Global Burden of Disease Collaborative Network},
|
||||
date = {2020},
|
||||
}
|
||||
@ -0,0 +1,18 @@
|
||||
|
||||
@article{ursu_drugcentral_2017,
|
||||
title = {{DrugCentral}: online drug compendium},
|
||||
volume = {45},
|
||||
issn = {0305-1048, 1362-4962},
|
||||
url = {https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkw993},
|
||||
doi = {10.1093/nar/gkw993},
|
||||
shorttitle = {{DrugCentral}},
|
||||
pages = {D932--D939},
|
||||
issue = {D1},
|
||||
journaltitle = {Nucleic Acids Research},
|
||||
shortjournal = {Nucleic Acids Res},
|
||||
author = {Ursu, Oleg and Holmes, Jayme and Knockel, Jeffrey and Bologa, Cristian G. and Yang, Jeremy J. and Mathias, Stephen L. and Nelson, Stuart J. and Oprea, Tudor I.},
|
||||
urldate = {2023-04-10},
|
||||
date = {2017-01-04},
|
||||
langid = {english},
|
||||
file = {Full Text:/home/dad/Nextcloud/Zotero_data/storage/7W6THRK6/Ursu et al. - 2017 - DrugCentral online drug compendium.pdf:application/pdf},
|
||||
}
|
||||
@ -0,0 +1,10 @@
|
||||
|
||||
@online{noauthor_fda_nodate,
|
||||
title = {{FDA} Drug Approval Process},
|
||||
url = {https://www.drugs.com/fda-approval-process.html},
|
||||
abstract = {It can take up to \$2 billion and 12 to 15 years to get a drug from the test tube to the market. What happens at the {FDA} to get this drug safely to you?},
|
||||
titleaddon = {Drugs.com},
|
||||
urldate = {2023-04-12},
|
||||
langid = {english},
|
||||
file = {Snapshot:/home/dad/Nextcloud/Zotero_data/storage/VTIGXXJB/fda-approval-process.html:text/html},
|
||||
}
|
||||
@ -0,0 +1,2 @@
|
||||
|
||||
cat ./*.bib > ../References.bib
|
||||
@ -1,34 +1,52 @@
|
||||
\begin{tikzpicture}
|
||||
\begin{pgfonlayer}{nodelayer}
|
||||
\node [style=emptyBox] (0) at (-18.5, 4) {Compound Safety};
|
||||
\node [style=emptyBox] (1) at (-18.5, -3.5) {Compound Effecacy};
|
||||
\node [style=Red Box] (3) at (-3.5, -4.25) {\begin{tabular}{l} Conclusion State \\ $\bullet$ Status\\ $\bullet$ Duration \\ $\bullet$ Enrollment \end{tabular}};
|
||||
\node [style=Red Box] (4) at (-2, 1) {\begin{tabular}{l} Snapshot State \\ $\bullet$ Status\\ $\bullet$ Duration \\ $\bullet$ Enrollment \end{tabular}};
|
||||
\node [style=Red Box] (5) at (3.25, -2) {Market Conditions};
|
||||
\node [style=Box] (6) at (-17.5, -5.5) {Sponsor Changes};
|
||||
\node [style=Box] (7) at (-15.25, 0.25) {\begin{tabular}{l}Prior \\ Trials \end{tabular}};
|
||||
\node [style=emptyBox] (8) at (-2.25, 4) {Beliefs about Compound};
|
||||
\node [style=emptyBox] (10) at (3.25, -5.25) {Unobserved};
|
||||
\node [style=Box] (11) at (3.25, -6.25) {Observed: Control};
|
||||
\node [style=Red Box] (12) at (3.25, -7.25) {Observed: Of interest};
|
||||
\node [style=emptyBox] (13) at (-11, 4) {Current Adverse Events};
|
||||
\node [style=emptyBox] (14) at (-10.5, -1.5) {Measured Effectiveness};
|
||||
\node [style=Box] (15) at (4.75, 1) {Disease Burden};
|
||||
\node [style=Red Box] (0) at (4, -1.5) {Will Terminate?};
|
||||
\node [style=Red Box] (1) at (-4.25, -1.5) {Market Measures};
|
||||
\node [style=emptyBox] (4) at (-6, -7.5) {Unobserved};
|
||||
\node [style=purple box] (5) at (0, 2) {Enrollment};
|
||||
\node [style=Red Box] (8) at (-5.75, -6) {Relationships of interest};
|
||||
\node [style=emptyBox] (9) at (12.25, 7.5) {Fundamental Efficacy/Safety};
|
||||
\node [style=emptyBox] (10) at (0, 9.25) {Previously Observed Efficacy/Safety};
|
||||
\node [style=emptyBox] (12) at (12.25, 0) {Currently Observed Efficacy/Safety};
|
||||
\node [style=Gray Box] (13) at (7, -6) {Observed, adjustment set 1};
|
||||
\node [style=GreyBoxDotted] (14) at (7, -7.25) {Observed, adjustment set 2};
|
||||
\node [style=GreyBoxDotted] (15) at (-4, 4.25) {Planned Enrollment};
|
||||
\node [style=GreyBoxDotted] (16) at (4.5, 4.5) {Anticipated Enrollment};
|
||||
\node [style=GreyBoxDotted] (17) at (7.5, 3) {Measured Enrollment};
|
||||
\node [style=Gray Box] (18) at (-6.5, 1.5) {Population};
|
||||
\node [style=GreyBoxDotted] (19) at (5.75, 1.5) {Current Status};
|
||||
\node [style=Gray Box] (20) at (0, 7.5) {Decision to procced with Phase III};
|
||||
\node [style=Gray Box] (21) at (0, -3.5) {Condition};
|
||||
\node [style=Gray Box] (22) at (14.5, -4.25) {Elapsed Duration};
|
||||
\node [style=purple box] (23) at (7, -8.5) {Partially observed};
|
||||
\end{pgfonlayer}
|
||||
\begin{pgfonlayer}{edgelayer}
|
||||
\draw [style=RightArrow] (4) to (3);
|
||||
\draw [style=RightArrow] (5) to (3);
|
||||
\draw [style=Light Arrow] (0) to (7);
|
||||
\draw [style=Light Arrow] (1) to (7);
|
||||
\draw [style=Light Arrow] (7) to (8);
|
||||
\draw [style=Light Arrow] (8) to (4);
|
||||
\draw [style=Light Arrow] (6) to (3);
|
||||
\draw [style=Light Arrow] (0) to (13);
|
||||
\draw [style=Light Arrow] (13) to (3);
|
||||
\draw [style=Light Arrow] (1) to (14);
|
||||
\draw [style=Light Arrow] (14) to (3);
|
||||
\draw [style=Light Arrow] (15) to (5);
|
||||
\draw [style=Light Arrow] (15) to (4);
|
||||
\draw [style=RightArrow] (5) to (4);
|
||||
\draw [style=lightredarrow] (1) to (0);
|
||||
\draw [style=lightredarrow] (1) to (5);
|
||||
\draw [style=lightredarrow] (5) to (0);
|
||||
\draw [style=RightArrow] (15) to (5);
|
||||
\draw [style=RightArrow] (5) to (16);
|
||||
\draw [style=RightArrow] (5) to (17);
|
||||
\draw [style=RightArrow] (5) to (19);
|
||||
\draw [style=RightArrow] (18) to (5);
|
||||
\draw [style=RightArrow] (18) to (1);
|
||||
\draw [style=RightArrow] (9) to (10);
|
||||
\draw [style=RightArrow] (10) to (20);
|
||||
\draw [style=RightArrow] (9) to (12);
|
||||
\draw [style=RightArrow] (12) to (0);
|
||||
\draw [style=Light Arrow] (21) to (1);
|
||||
\draw [style=Light Arrow, in=-75, out=105] (21) to (15);
|
||||
\draw [style=Light Arrow] (21) to (0);
|
||||
\draw [style=Light Arrow, in=-120, out=165, looseness=2.00] (21) to (18);
|
||||
\draw [style=Light Arrow, in=180, out=180, looseness=2.75] (21) to (20);
|
||||
\draw [style=Light Arrow, bend right=60, looseness=1.75] (21) to (9);
|
||||
\draw [style=Light Arrow] (21) to (5);
|
||||
\draw [style=lightredarrow, in=135, out=45, loop] (8) to ();
|
||||
\draw [style=RightArrow, in=-180, out=165, looseness=2.00] (1) to (20);
|
||||
\draw [style=RightArrow, in=120, out=180, looseness=1.25] (10) to (15);
|
||||
\draw [style=RightArrow] (20) to (5);
|
||||
\draw [style=RightArrow] (18) to (15);
|
||||
\draw [style=RightArrow] (22) to (0);
|
||||
\draw [style=RightArrow] (22) to (5);
|
||||
\end{pgfonlayer}
|
||||
\end{tikzpicture}
|
||||
|
||||
Loading…
Reference in New Issue