JobMarketPaper/Paper/sections/05_LitReview.tex

\documentclass[../Main.tex]{subfiles}
\graphicspath{{\subfix{Assets/img/}}}

\begin{document}

% Outline
% - Introduce and frame problem
%   - Phases & regulatory part
%   - Large number of failures at each phase
%   - There are multiple ways to measure this
%   - Estimation of failures at phase and failures per development path
%   - Talk about impact of making these closer together
% - Trying to develop more by tweaking external world:
%   - Pull incentives 
%   - Increase in market sizes.
%   - Uncertanty in Intellectual Property
% - Understanding failure modes
%   - EK and Hwang
%   - discuss missing section of operational concerns
%   - Introduce metabio
% - Once again bring up my work here.
%   - 
%   - 

\subsection{Drug development process and failure rates}
% Abrantes-Metz, Adams, Metz (2004)
% - What correlates with successfully passing clinical trials and FDA review?
% - 
\cite{abrantes-metz_pharmaceutical_2004} 
described the relationship between
various drug characteristics and how the drug progressed through clinical trials.
This descriptive estimate used a 
mixed state proportional hazard model and estimated 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 and II trials, while Phase 3 trials generally have a higher rate of 
success than failure after 91 months.

%DiMasi FeldmanSeckler Wilson 2009
\cite{dimasi_TrendsRisks_2010} examine the completion rate of clinical drug 
develompent and find that for the 50 largest drug producers, 
approximately X\% of their drugs under development
\todo{FILL IN X}
successfully completed the process. 
They note a couple of changes in how drugs are developed over the years they 
study (clinical development started between 1993 and 2004). 
This included that drugs began to fail earlier in their development cycle in the 
latter half of the time they studied. 
This may be an operational change to reduce the cost of new drugs. 

\cite{dimasi_ValueImproving_2002}
used data on 68 investigational drugs from 10 firms to simulate how reducing
time in development reduces the costs of developing drugs. 
He estimates that reducing Phase III of clinical trials by one year would 
reduce total costs by about 8.9\% and that moving 5\% of clinical trial failures
from phase III to Phase II would reduce out of pocket costs by 5.6\%. 


% 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
% Don't think this is applicable.

\subsection{Market incentives and drug development}
%%%%%%%%% What do we know about drug development incentives?
\subsection{What do we know about drug development incentives?}
% Introduce section
% -  Dranov et al 2022 - demand pull seems to bias follow up drug development.
%      - increasing demand doesn't necessarily result in new compounds (check this). Risks.
\cite{dranove_DoesConsumer_2022} examined whether increased demand for drugs 
will increase the development of novel drugs. 
Using measures of the scientific novelty of drug compounds after the creation
of Medicare part D, they found that most development occurred in the least
novel categories of drugs, in spite of a relatively constant growth in novel 
compounds. 

\cite{dranove_DoesConsumer_2022} use the implementation of Medicare part D 
to examine whether the production of novel or follow up drugs increases during 
the following 15 years. 
They find that when Medicare part D was implemented -- increasing senior 
citizens' ability to pay for drugs -- there was a (delayed) increase 
in drug development, with effects concentrated among compounds that were least
innovative according to their classification of innovations.
They suggest that this is due to financial risk management, as novel 
pharmaceuticals have a higher probability of failure compared to the less novel
follow up development. 
This is what leads risk-adverse companies to prefer follow up development.



%   -  acemoglu and linn 2004 - population size matters.
%   - Population ties into the number of drugs available, and operational (recruitment) concerns
%   - In general, there are going to be many confounding variables.
% - 
% - Exogenous demographic trends has a large impact on the entry of non-generic drugs and new molecular entitites.
On the side of market analysis, 
\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.


% Cerda 2007 - Endogenous innovations in the pharmaceutical industry
% from abstract %TODO: Read better
% Market size, population, and existence of drugs are endogenous
% from the abstract I get the impresssion that it is:
% - large population -> large market -> more profitable -> more drugs
% - more drugs -> better survivability -> larger market 
% Applicable because: Need to separate population and market effects.
% Does this mess with my results? I don't think so because of the relatively short time in trials. Not enough time to effect population back, but it might have another effect.
\cite{cerda_EndogenousInnovations_2007} 
suggests a two-way, long term relationship between market size and drug 
development.
They suggest that a large population with a condition implies a (relatively) 
larger market, which improves the profitabilty and thus number of drugs with that
condition. 
Then the drugs improve mortality, increasing the relative population.
They do find evidence of the impact of both population and market size
on the creation of new drugs. 


% van der gronde et al 2017 Addressing the challenge of high-price prescription drugs
% Massive number of policies used to try to reduce costs. These will affect production decisions. 
% Some of the unintended consequences of that (in terms of reduced development incentives) include
% - reducing development costs - side effect of lower quality evidence
% - Preference policy (e.g. policies about using generics first etc) - side effect of shorter life cycle for patented (novel) drugs.
% - these are focused on reducing expenditures, i.e. they reduce profit. Some of them feed back into the development process.

\cite{vandergronde_AddressingChallenge_2017}
documents many of the things driving drug development choices. 
\begin{itemize}
    \item Policies that encourage low cost generics shorten the life cycle of
        patented/novel drugs.
    \item Some diseases have lower safety and efficacy standards applied to them 
        compared to similar diseases. These tend to have higher R\&D due to the 
        lower costs involved.
    \item As much of the "low hanging fruit" in drug development has been developed,
        R\&D expenses have been increasing.
\end{itemize}


% Dubois et al 2015 - Market Size and pharmaceutical innovation
% estimate the relationship between marekt size and the innovation in pharmaceuticals
% elasticity of innovation w.r.t. expected market size of 0.23, thus $2.5 billion in 
% market size required to get a new chemical entity.
\cite{dubois_MarketSize_2015}
examined the ``elasticity of innovation'', i.e. the ``additional revenue required
to support the invention of a new chemical entity.''
They found that a marginal drug will require approximately a \$2.5 billon increase
in expected revenue.

% Gupta
% - Inperfect intellectual property rights in the pharmaceutical industry
\cite{gupta_OneProduct_2020} discovered that uncertainty around which patents
might apply to a novel drug causes a delay in the entry of generics after 
the primary patent has expired. 
She found that this delay in delivery is around 3 years. 


\subsection{What do we know about how Clinical Trials operations?}
%interview with Adam George
% - clinical trials are often handled by contractors
% - they plan sites, start times, etc from beginning.
% - Running late is normal.
In a personal interview with someone who works for a company that runs clinical
trials, I learned about how clinical trials will typically proceed. 
\todo{Figure out best way to cite this}
\begin{itemize}
    \item Quote a job (one side of company): N, timeline, etc
    \item Allocate resources (sites, doctors, etc) to try to accomplish
    \item Sales vs Operations conflict, leading to lateness/issues delivering, etc.
\end{itemize}

% Bess Stillman - look at difficulties joining oncology trials

% Random sample of Clinicaltrials.gov - how many closed due to operational problems?
% TODO: random sample 171, about 30% mentioned recruitment issues

% Results on enrollment projection
% - nothing really good exists.
% - Multiple models, no comparison.
% - no cross validation, only tested on  a few trials.

% Thus we should look at the effects that operational concerns have.

\subsection{Understanding Failures in Drug Development}

% DISCUSS: Different types of failures
There are myriad of reasons that a drug candidate may not make it to market, 
regardless of it's novelty or known safety.
In this work, I focus on the failure of individual clinical trials, but the 
categories of failure apply to the individual trials as well as the entire
drug development pipeline.
They generally fall into one of the following categories:
\begin{itemize}
    \item Scientific  Failure: When there are issues regarding 
        safety and efficacy that must be addressed. 
        The preeminient question is: 
        ``Will the drug work for patients?''
        %E.Khm, Gupta, etc.
    \item Strategic Failure: When the sponsors stop development because of 
        profitability
        %Whether or not the drug will be profitiable, or align with
        %the drug developer's future Research \& Development directions i.e.
        ``Will producing the drug be beneficial to the 
        company in the long term?''
        %E.Khm, Gupta, GLP-1s, etc.
    \item Operational concerns are answers to: 
        %Whether or not the developer can successfully conduct
        %operations to meet scientific or strategic goals, i.e. 
        ``What has prevented the the company from being able to 
        finance, develop, produce, and market the drug?''
\end{itemize}
It is likely that a drug fails to complete the development cycle due to some
combination of these factors. 


%USE MetaBio/CalBio GLP-1 story to illuistrate these different factors.
\cite{flier_DrugDevelopment_2024} documents the case of MetaBio, a company
he was involved in founding that was in the first stages of
developing a GLP-1 based drug for diabetes or obesety before being shut down
in . 
MetaBio was a wholy owned subsidiary of CalBio, a metabolic drug development 
firm, that recieved a \$30 million -- 5 year investment from Pfizer to 
persue development of GLP-1 based therapies. 
At the time it was shut down, it faced a few challenges:
\begin{itemize}
    \item The compound had a short half life and they were seeking methods to 
        improve it's effectiveness; a scientific failure.
    \item Pfizer imposed a requirement that it be delivered though a route
        other than injection (the known delivery mechanism); a strategic failure. 
    \item When Pfizer pulled the plug, CalBio closed MetaBio because they 
        could not find other funding sources; an operational failure.
\end{itemize}

The author states in his conclusion:
\begin{displayquote}
    Despite every possibility of success, 
    MetaBio went down because there were mistaken ideas about what was 
    possible and what was not in the realm of metabolic therapeutics, and 
    because proper corporate structure and adequate capital are always 
    issues when attempting to survive predictable setbacks.
\end{displayquote} 

From this we see that there was a cascade of issues leading to the failure to 
develop this novel drug. 

% NOW discuss efforts to measure the impact of different aspects

\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.


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}).
% causal separation of strategic exits etc.

% I don't think I need to include modelling enrollment here. 
% If it is applicable, it can show up in those sections later.



\end{document}