MMS Holdings Blog

ACRP Presentation: An Operational Plan for Clinical Trial Disclosures

Posted on Thu, May 21, 2015 @ 01:46 PM

joeBy Dimple Kharwar, Manager, Business Development


Joe Archer, Associate Director of Clinical Trial Disclosures at MMS Holdings, was in attendance at the annual Association of Clinical Research Professionals (ACRP) Global Conference & Exhibition in Salt Lake City, Utah on April 25-28th, 2015.   The ACRP is the largest annual conference primarily focused on the conduct of clinical trials.  The conference covers topics such as training, Quality Assurance, clinical trial operations, and clinical trial technologies, and the target audience is clinical research professionals from across industry (clinical sites, pharmaceutical companies, and CROs, who support trial recruitment, monitoring, trial management, etc.).  The conference offers over 100 educational sessions and numerous opportunities for networking.


Joe presented “Disclosure Requirements: A Moving Target” as part of the Emerging & Specialized Topics educational track.  In his presentation he laid out an operational plan using existing technology to help organizations manage the intricate workflow associated with trial disclosures.  As a background for the talk, Joe covered the current state of compliance in the US, along with the evolving regulations within the US, EU, and across the globe that are set to expand reporting requirements.   


Sharing his experience, Joe said, “There are many presentations being given that describe the complexities surrounding Trial Disclosure Laws, but very few of these discussions provide solutions. The ACRP was a great setting to provide the audience an operational blueprint for managing disclosures.”  The session was well received by attendees, and initial feedback indicated that the topic generated interesting discussions and dialogue amongst the attendees. 


If you’re interested in more information about Clinical Trial Disclosure Reporting at MMS, please contact, or call (734) 245-0310.


Trial Disclosure at MMS:


The MMS Disclosure Services team is your ideal strategic outsourcing partner for clinical trial disclosures. Our Staff Expertise, Proven Processes and Exclusive Technology (TrialAssure™), ensure quality, guarantee compliance, and provide comprehensive disclosure reporting - all in a centralized, cost effective manner!


TrialAssure™ Features and Benefits:


TrialAssure™ is a proprietary software reporting tool that facilitates workflow management. It ensures quality, compliance, and completeness of clinical disclosure reporting, while serving as a centralized and definitive (single source) catalogue across an organization’s drug development pipeline.


Tags: trial disclosure

PharmaSUG 2015: MMS Attends the May 17-20th Conference in Orlando, FL!

Posted on Wed, May 13, 2015 @ 09:15 AM

By Shannon Plasters, Alliance & Strategic Partnerships Lead

pharmasugMMS will be in attendance at the PharmaSUG Annual Conference in Orlando, Florida on May 17-20th  to present a poster and deliver an educational presentation, as well as attend the workshops, demonstrations, and networking events. PharmaSUG is the Pharmaceutical Industry SAS Users Group whose mission is to provide a forum for the exchange of information and the promotion of new ideas related to software and tools (including the SAS programming language) in the pharmaceutical industry.  MMS will be sending several representatives from the Programming and Biostatistics functional areas to participate in the event, share their experiences, and learn cutting-edge ideas from their colleagues at this premier annual conference for SAS users. 

The conference will offer workshops, posters, and presentations across a wide range of topics, including Data Standards, Applications Development, Submissions Standards, Industry Basics, and Career Planning.   Christopher Hurley, Senior Manager, Clinical Programming and Biostatistics will be co-presenting an exciting talk to provide sound advice to SAS programmers to help them develop their career and cultivate their leadership potential.  “I am passionate about this topic, and love to help SAS programmers develop solid gold skills and advance in their careers.  As a manager, it’s one of the most rewarding things about my job,” said Chris.  “I’m looking forward to sharing my insights with the attendees at PharmaSUG.”

Harry Haber, Senior Biostatistician, will be presenting a poster describing analysis methods for an adaptive trial design called Sequential Parallel Comparison Design (SPCD).  There are some important benefits of this study design if implemented properly, including increasing the number of subjects who receive active treatment, and improving the statistical power of the study.  “Please stop by the poster session (Poster 23) to learn about SPCD adaptive study design,” said Harry.  “Our poster will compare and contrast 3 different analysis methods (and the SAS code used to implement these methods) that will assist anyone who is considering designing a clinical trial in this model.”

James Zuazo, Associate Manager, Biostatistics, is looking forward to networking with industry colleagues and attending the educational sessions on offer.  “SAS is such an important part of our industry, and I’m eager to share my expertise as well as learn new tips and tricks that will help me on a day-to-day basis.  As an organization, MMS has learned a lot about SAS functionality and how to share that expertise internally; I anticipate expanding my knowledge base even further with other colleagues in our industry.”

As an everyday user of SAS, MMS fully understands the importance of this software to the development of new drugs, and values the opportunity to collaborate with other SAS users across the industry.  MMS is committed to continuing to support the mission of PharmaSUG and is excited about attending this year’s event. This year’s PharmaSUG attendees from MMS look forward to seeing you in Orlando! 


Tags: PharmaSUG 2015

6 Features of a Statistical Analysis Plan that Can Improve Efficiency in the Programming Process

Posted on Tue, May 05, 2015 @ 09:00 AM

By Karen Rosales, Biostatistician

A well written Statistical Analysis Plan (SAP) is a critical component of successful data analyses in clinical trials. The SAP further elaborates on the main features of the analyses described in the protocol, including a detailed description of how the statistical analysis of the primary, secondary and other endpoints should be executed. The SAP is an official regulatory document that represents an agreement between Clinical and Statistics regarding what analyses should be performed to address all the clinical study objectives. This is the document that programmers use for creating analysis dataset specifications and developing tables, listings and figures (TLFs), alongside the TLF shells (which should closely follow the SAP).

Six important features of a SAP that can facilitate the ease of generating outputs from a programming point of view include:

  1. Statistical Methods to be Used

The statistical methods of the SAP are very crucial and should be described properly as the results from the tests using these methods will address the objectives of the study. This section should detail how the analyses for the study should be done and how the data should be summarized for the study subjects. Summary statistics (like mean, standard deviation, percentiles, counts, percentages) should be stated as ways of summarizing the data when appropriate. Statistical analyses to be used (such as t-test, ANOVA, cox regression) should be described in a way that programmers can easily translate into programs for analysis. For more complex analyses, having a sample SAS code would be helpful.

2. Study Endpoints

Each endpoint should be well-defined. A study endpoint refers to the outcome that the clinical trial is trying to measure to determine whether the drug or intervention being studied is beneficial. Some examples of endpoints include relief of symptoms, improvement of quality of life, and difference in survival between the treatment and the control group. The SAP should layout how each endpoint will be measured. An endpoint measurement could mean a single value, a specific category, a total, the mean, etc. The SAP should clearly state which measurement will be used for each endpoint and how to arrive at that measurement, especially when it calls for something more complex than getting the sum or mean.

3. Data Handling Rules

The SAP should contain imputation rules, which describe how to handle missing data or algorithms for derived variables. This could refer to missing dates, unanswered items in a questionnaire, or missed visits/assessments. These rules are set up to avoid the bias of over- or under-estimation and to maximize the usability of whatever data are available, and are therefore very important for a meaningful analysis later.

4. Analysis Populations

The SAP should define the analysis populations to group subjects together according to set criteria. These could be an intent-to-treat or full analysis set, a safety set, a per-protocol set, or other sets. Defining these populations is critical for running the efficacy and safety analyses of the study with the least amount of bias possible.

5. Well-defined Analysis Visit Windows

Subject visits, though clearly scheduled in the protocol, often do not occur on the actual protocol-scheduled days. Subjects could miss a visit or have multiple visits in one ‘visit window’. To anticipate late, unscheduled, or multiple visits and still have a meaningful efficacy and/or safety analysis, well-defined analysis visit windows should be planned whenever possible. Some studies use nominal visits which are the observed visits; this should be clearly explained in the SAP as well. But for studies where analysis visit windows are feasible, they are useful to utilizing all data. As much as possible, all the data should fall into one of the visit windows in the study. This section should also state the rule for handling multiple data falling in the same analysis visit window, e.g. only choose the latest assessment. Having these rules clearly stated beforehand is very helpful in running the outputs for analysis later.

6. Clearly Defined Meaning of Baseline

In principle, baseline is usually the last non-missing value/assessment before the first day of study drug. There could be other versions of this definition, depending on endpoints or study design (for example, studies with different stages). A clear definition is helpful to the programming team when creating consistent outputs across the study.

There are other important elements that need to be further elaborated in the SAP, but they fall under the ‘standard’ categories which are already noted in the protocol, like concomitant medications, adverse events, etc.  The 6 elements described above are necessary to have a clear understanding of how to build the analysis datasets and TLFs. Clear definitions within the SAP on all aspects of the analyses variables help to ensure consistency and accuracy of all data results and deliverables, including analysis datasets and TLFs.

Tags: Statistical Analysis Plan

The Critical Role of a Regulatory Operations Specialist in a Successful Submission: 10 Key Success Factors

Posted on Wed, Mar 25, 2015 @ 09:58 AM

ROS_submissionsBy Rashmi Dodia, MS – MMS Regulatory Operations Specialist

For flawless submissions rooted in quality from the early stages of conception through successful execution and submission, a Regulatory Operations Specialist (ROS) can be considered the Captain of the Ship. Their role is not as simple as many consider it to be. A submission is not about attaching pre-written documents to an email and hitting the “Send” button on the FDA Gateway. It involves strategy and efficient operational skills for compilation, and minimizing the chances of errors that can involve tremendous amounts of rework.

One could liken the role of a ROS to the post-production responsibilities of a movie producer – they need to make sure the final submission is a polished and a truly finished product!  They are the ones that put all the pieces together, make sure everything is in the right place per the Agency requirements, ensure all functionalities of the final output xml or PDF work, and present a true submission “package.” They are the last, critical step in finalizing a submission. Their key tasks include laying out a detailed project plan for their submission in accordance with the eCTD structure, publishing submission-ready documents, compiling and publishing submissions using a publishing tool, corresponding with the Health Agencies (HA) about eCTD publishing specifications, and oversight of document management for effective tracking, versioning and archiving of regulatory documents. All of these tasks require immense patience, the ability to multitask, and a sound knowledge of regulatory requirements and expectations of the concerned HA.  Even a minor oversight on the part of an ROS can lead to submissions being rejected.

Here are 10 key attributes of an experienced ROS:

1)      Knowledge of regulatory modules and structure: ROS’s start their work with understanding the type of application they are working on. Their deliverables can range from FDA IND and NDA applications/amendments to CTAs or MAAs in different parts of the world. They have to dig deep and chalk out an action plan.

2)      In-Depth Knowledge of CFRs and Guidances: Just having the knowledge of application types and procedures is not enough; use of CFRs and guidances is an indispensable part of the ROS’s daily routine. The more familiar they are with the regulations, the easier it is to reference them. All work on an application has to be compliant with the guidance at every stage.

3)      Attention to Detail and Quality: Once they have charted out a plan conducive to their submission type, the next step is to ensure all the required documents are procured from the right sources and published according to submission-ready standards.  A ROS has to have a keen eye for detail, and ensure that the formatting and finalization of documents is done as per client and internal quality standards. It may require conducting quality checks on a document at different stages to ensure the highest level of quality is met.

4)      Effective Document Management: In the process of putting together a submission, there is a lot of back and forth of documents between the client, external vendors, strategists, medical writers, and document specialists. In such a scenario, a ROS has to ensure proper versioning of documents through effective document, including incorporating additional updates/ last minute changes. The ROS also has to make certain the deliverables stay on track. They should have the ability to speak up, and more importantly, they must ensure the job gets done.

5)      Expert Knowledge of the Publishing Tool: This is one of the more important tasks of a ROS on a daily basis. There is a huge learning curve involved in this area. A ROS has to learn how to import documents into the publishing system. They have to compile the submission patiently, and that involves a lot of guidance referencing in order to place the components in their correct module in the submission outline, name them as per FDA conventions, version them appropriately, select the right lifecycle operators for documents, and manage datasets and program files.  When finally everything seems about right, they can publish the submission, expecting the least number of errors. But on occasion, they are in for a surprise as the validation tool is programmed as per FDA guidances to catch anything that seems out of place.  The important thing is to submit a quality submission to the HA in time.

6)      Effective Communication with HA: At times there may be questions regarding the HA preferences for certain document types or placement of documents in the eCTD outline, or other concerns like datasets exceeding the FDA specified size limit. In these unexpected situations, the wisest thing to do is contact the Agency with questions and concerns and take action accordingly.

7)      Thinking Outside of the Box: Although regulations and guidances provide a road map, there are “gray” areas that require interpretation in almost every major submission. There are instances when the “go-to process” will not work, and a ROS is required to creatively think outside of the box to successfully complete the project. Additionally, it is a prime responsibility of a ROS to document such instances and experiences for future reference.

8)      Streamline and Improve Current Processes: The learning curve never ends for a ROS and they always have to be open to new challenges and opportunities. One of their main responsibilities is to create workflows, SOPs, and guidances for newer projects that can be referenced in the future for similar work. They need to keep streamlining and improving on the current methods and processes as they continue to learn better ways of doing things.

9)      Keeping Track of Submission Timelines: A ROS has to keep a track of all the submission activities and ensure they are on top of their game. They need to have timely communication with the strategists and clients to ensure everything is moving along smoothly and none of the submission timelines are getting missed. Knowledge of regulatory timelines and framework is a key factor in ensuring all required documents and forms reach the agency on time.

10) Awareness of Client Expectations: It is important to meet the HA timelines as well as meet client expectations. A ROS has to ensure their job is well done, take into consideration and accommodate client requests (even if they are last minute), and look for ways to provide excellent service to surpass their expectations. Wowing the customer with timely support is a part of their daily responsibilities. Looking for ways to better support clients is a vital part of the MMS business model. Since Regulatory Operational tasks usually begin when the other departments have completed their work, the final responsibility of a successful submission is carried on their shoulders

Tags: regulatory submissions, Regulatory Team

10 Ways to Make Friends with ADaM (Datasets)

Posted on Wed, Mar 11, 2015 @ 08:30 AM

By Justin Sjogren, Senior Biostatistician

Who is the new face of analysis dataset reporting in the pharmaceutical industry?  It’s ADaM (Analysis Data Model), a standardized dataset structure brought to us by CDISC and fully laid out for us in the Analysis Data Model Implementation Guide. ADaM datasets are the next step following SDTM (Standard Data Tabulation Model), which are datasets that display raw data in an organized, CDISC-compliant structure.

So how else would we describe ADaM? Well, he’s a strong, powerful figure who represents organization, hard work, and a bridge to something great. Even though he hasn’t been around long, his conventions have led to many success stories and have proven to stand the test of time.

Admittedly, his complex nature makes it tough for many of us to understand him, and sometimes the effort involved in getting to know him makes us wish we were making friends with someone else. But, I assure you if you take the time to get to know him and study his tendencies, you will be glad to walk away with a new friend.

But don’t forget to follow protocol when you start working with him; after all, he’s like a member of the royal family! Always keep your ADaM guidelines in mind (see, but below is a list of some additional things you can do to help facilitate successful and high-quality analysis. I hope you find these tips and tricks useful for your next ADaM project!

  1. Analysis Day (ADY) should be re-calculated
    • It may be tempting to bring in your --DY variables directly from SDTM into ADY, but this can be dangerous.
    • If your reference start date (RFSTDTC) and treatment start date (TRTSDTM) are inconsistent, this leads to differences between --DY and ADY, which can throw a wrench in your analyses.
    • It’s good practice to re-calculate study day based on treatment start date and assessment date.
  2. The importance of analysis visit (AVISIT) should not be overlooked
    • Consult your statistician and ensure this is defined clearly in the SAP.
    • If visit windows are used, ensure they are contiguous and it is clear what to do if multiple visits occur within the same window.
    • Don’t forget about those unscheduled visits!
    • Your visit-based domains and results depend upon AVISIT.
  3. QC your ADaM specifications against your SAP
    • Even though they are in a different format, they should “say” the same thing.
    • Have a statistician perform a consistency check between both of these documents because any differences will lead to confusion and (possibly) errors.
    • Start with identifying areas from the SAP that should be in ADaM, and start checking!
  4. Always triple check the important dates
    • Keep in mind the dates that matter most, such as first dose date (TRTSDTM) and last dose date (TRTEDTM).
    • Items like randomization date and date of last dose in double-blind can also be very critical.
    • Treatment-emergent adverse event (TEAE) counts are based on these dates, so one incorrect date and your TEAE counts change (fingers snapping) “like that.”
    • Baseline is also dependent on these dates, so an error in these dates can lead to a different baseline value.
  5. Include all randomized subjects in your Basic Data Structure (BDS) domains
    • It’s tempting to only include subjects who received treatment (e.g., safety set) in your BDS domains. But the inevitable “oh no” may happen when the FDA requests a table that was originally run on the safety population to be run on ITT.
    • If you don’t put all ITT subjects in ADaM, you’ll likely have to retrieve some from SDTM and then update ADaM to include them, and I suspect you’ll want to avoid that.
  6. Include a row for baseline
    • Don’t omit the baseline row just because you see the BASE variable
    • The characteristics about the baseline record, such as date, are still going to be needed.
    • Plus, the structure is useful for most PROC steps.
  7. Know the contents (populations, records) that are included in each ADaM dataset
    • For example, does AEAE include only TEAEs or non-TEAEs as well? How about double-blind vs. open-label AEs?
    • What populations are included in ADVS (vital signs), ADEG (ECGs) and ADLB (labs)? All randomized or only those receive study drug?
  8. Include ALL levels of MedDRA and WHO Drug in ADAE (adverse events) and ADCM (concomitant medications)
    • Don’t just include SOC (system organ class) and PT (preferred term), because you never know when your sponsor will want to review all levels of MedDRA coding.
    • Before you know it, Lower Level Term is in your table shell and you do not want to have to get it from SUPPAE (supplemental AE).
  9. Avoid the raw and input from SDTM
    • Try not to develop any ADaM datasets from raw data.
    • Skipping SDTM means any raw data used would have to be submitted to the regulatory agency (e.g., FDA).
  10. Invest your time in developing ADaM datasets
    • Let’s face it, ADaM programming takes a long time.
    • This is where your populations, derivations, imputations and calculations reside.
    • So time spent here is value-added, and can dramatically save time for your TLG programming.

Remember that high quality ADaM datasets will lead to quicker and more efficient TLG programming, smoother submissions, and fewer questions from agencies!

Tags: data management best practices

MMS at PhUSE Conference March 15 – 17th in Silver Spring, MD!

Posted on Tue, Mar 10, 2015 @ 07:02 PM

By Shannon Plasters, Alliance & Strategic Partnerships Lead

MMS will be attending the PhUSE Computational Science Symposium (CSS) conference in Silver Spring, Maryland on March 15-17th to present several posters, host a promotional booth, and attend the educational sessions.  PhUSE brings together Regulators, Industry, Technology, and Service Providers to address computational science needs in support of regulatorPhUSE Logoy review. Representatives from several MMS functional departments will be in attendance, including Biostatistics, Programming, Clinical Trial Disclosures, and Project Management.  MMS attendees are very excited to participate in the event, learn from other attendees, and share their experiences related to the access and review of data to support pharmaceutical product development.

Chris Hurley, Senior Manager, Clinical Programming and Biostatistics at MMS, was appointed to  the PhUSE Board of Directors in 2015, and has been involved in the planning and organizing of this year’s event.  “As part of the planning committee, I’ve seen a huge effort by the PhUSE organizers to put this event together.  Chris Decker and his team have done a fantastic job, and I know the attendees will gain a lot of insight from their participation in the working groups, educational sessions, and networking events at the CSS!”

MMS Holdings has been an excellent supporter of PhUSE.  In addition to exhibiting this year at the CSS, MMS has provided volunteers for many PhUSE events including speakers, a co-chair, workgroup participants, project leads and director level support.  PhUSE provides opportunities for collaboration between industry, academia, and agencies to advance clinical research, and this takes the combined input of many volunteers and sponsors.  MMS understands the importance of this work and is fully committed to supporting PhUSE in these opportunities. 

Chris will also be presenting a poster that represents outputs from the PhUSE Standard Scripts Working Group. “Please meet me at the poster session, Monday at 5pm, to learn about the progress this team has made toward standardization of code and typical data displays for analysis and reporting.  Our Standard Scripts are the product of a collaborative effort from volunteers at PhUSE, industry, and the FDA.  Come see what these scripts are all about and consider how you may be able to use them, and possibly even volunteer to support this team.  We love volunteers!”

The MMS Clinical Trial Disclosures team will also be presenting a poster on efficient workflows in the data disclosures arena.  Joe Archer, MMS Associate Director in Clinical Trial Disclosures and key developer of MMS’ proprietary and leading disclosure tool TrialAssure, is looking forward to attending this year’s event and presenting new ideas in this emerging field. “The information I’ll be presenting will help life science companies achieve compliance when reporting to government registries and results databases, such as  MMS is glad to be a part of the PhUSE ongoing discussion related to access and review of data in support of product development.”

Please come by and visit the MMS booth if you plan on attending the PhUSE CSS!  MMS will be showcasing our expertise in ADaM datasets and sharing tips on achieving a high quality dataset for any submission.   

Tags: PhUSE Conference

Race & Ethnicity in Clinical Research

Posted on Wed, Mar 04, 2015 @ 01:26 PM

By Evie M Delicha Ph.D

Defining the Problem 

The role of race and ethnicity in biomedical research is constantly under debate. After sequencing the human genome and mapping human genetic variations, there are contradictory beliefs and practices about the use of race and ethnicity in clinical research. These beliefs and practices endorse the concept of a socially constructed race and ethnicity, yet at the same time they are often treated as genetic variables. Is it scientifically correct to treat race and ethnicity as genetic variables and draw conclusions about treatment response, pharmacokinetic profiling, or disease prevalence?

Race and Ethnicity as a Social Construct

Race is defined as a category or group of people having common hereditary traits, whereas ethnicity is based on a shared cultural heritage. The notion of race was first adopted by the Europeans in the 16th century, when they used the phenotypic characteristics to define physical divisions among humans according to hereditary characteristics as reflected in morphology, and roughly captured as Black, White, and Asian (or Negroid, Caucasoid, and Mongoloid). They used this separation to classify the different groups of people they encountered through continental exploration.

Nowadays, the definition of race endeavors economic, political-social and cultural practices. Countries (especially these with population diversity) are gathering data based on race and ethnicity in order to monitor civil rights, unemployment, health care services, and educational opportunities. In the USA, the Office of Management & Budget (OMB) issued the Race and Ethnic Standards for Federal Statistics and Administrative Reporting that are set forth in Statistical Policy Directive No. 15. The OMB proposed a minimum of five race categories: American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, and White. As clearly stated in the Federal Register Letter (Oct 1997), these categories are viewed under the social perception of race in order to enforce civil rights laws:

 “….The racial and ethnic categories set forth in the standards should not be interpreted as being primarily biological or genetic in reference. Race and ethnicity may be thought of in terms of social and cultural characteristics as well as ancestry…..”

The proposed OMB race categories were also adopted by the FDA for reporting in clinical trials. In Europe, the EMA is not suggesting any specific race categorization, but it views race as an intrinsic ethnic factor (ethnicity has a broader meaning than race).  The EMA adopts their definition from the ICH E5:

“Intrinsic ethnic factors are factors that help to define and identify a subpopulation and may influence the ability to extrapolate clinical data between regions.”


The Geographic Barrier Reflects Fundamental Differences between Races and Ethnicities 

Despite the fact that regulatory agencies are supporting the social definition of race and ethnicity, these variables were employed not only in epidemiology, but also in the reporting of clinical trial outcomes. Therefore, the obscure properties of race and ethnicity as genetic variables should be investigated. Is a person’s cultural or ancestral background a key determinant in clinical response?

In the educational material distributed by the Human Genome Project in 2001, it was stated that “two random individuals from any one group are almost as different (genetically) as any two random individuals from the entire world.” This statement essentially refutes the supposition that racial divisions reflect fundamental genetic differences. The quantification of similarities within or between population clusters mainly depends on the level of genetic information analyzed, as well as the statistical norms employed to quantify them. Further research of more detailed genetic profiling indicated that the Ancestry-Informative Markers (AIMs) exhibit substantial diversity in frequencies between different race and ethnicity groups, and this diversity reflects geographic adaptation. Geographic barriers to gene flow have generated noticeable genetic differentiation between groups of different ancestry. In that way, there is an apparent correlation between gene frequencies and geography.

The example of the Ashkenazi Jewish women provides further data to support the argument that geographic adaption is responsible for differences between populations (rather than race and ethnicity).  The Ashkenazi Jewish women currently live all around the world, but they originally came from the same geographic regions (Germany, Russia, and Poland).  They share the same AIMs, and as a group, are at a greater risk of breast cancer than the general population.

Such observations became the fundamental assumption stimulating the emergence of personalized medicine. Let’s not forget the the efficacy of isosorbide dinitrate plus hydralazine (BiDil) in African-Americans (which led to a unique and controversial approval from FDA). A few years later, the GRAHF trial showed that it was the -344 T/C promoter polymorphism of CYP11B2 that influences the clinical outcome, and this polymorphism is more common in African American patients. As another example one could think of sickle cell anaemia, which is observed with high frequency among those of African and Mediterranean ancestry. Following this first observation, it was found that a mutation in the HBB gene that arose in Africa is responsible for the presence of the disease, and therefore any person with this mutation can develop sickle cell anaemia.

Reporting of Race and Ethnicity in Clinical Trials

As previously discussed, race and ethnicity are defined in social terms. There is no biological or genetic rationale behind the definition of race and ethnicity. Nevertheless, let’s suppose that race and ethnicity information is essential in clinical trials, and let’s suppose that these terms are used as a proxy for the geographic ancestral origin which may alter the clinical benefit. If that’s the case, then why there is no homogeneity in reporting such data in published literature? Although reporting of race and ethnicity has increased during recent years in study publications, there is no uniform terminology employed, leading to a possible misinterpretation of factors affecting health outcome. It is unclear how to interpret differences in study outcomes across different countries, cultures, and health systems; and the complexity resulting from this admixture is a reason for additional ambiguity. Currently, there is an enforcement of existing journal policies on how populations are defined and characterized in terms of race and ethnicity, and the relevance of these variables to the study outcome. However, still these guidelines are vague, not specifying smaller geographical divisions. Geographic adaptation is often ignored in large clinical trials (Japanese and Indians might both be considered Asians). The same level of enforcement should be utilized by health authorities and regulatory agencies in order to update the current reporting standards, and restricting the terms of race and ethnicity in clinical trials only to demographic information and not applying them to clinical outcome. Any variation of clinical outcome should be viewed under the spectrum of ancestry which is not related to the social definition of race.  Additionally, the regulatory agencies should move towards defining how genetic information should be analyzed in clinical trials, since genetic differences may alter the therapeutic result. The need is straight forward due to the increasing globalization of clinical trials, resulting in greater population diversity.



  1. Sankar P, Cho MKMonahan K, Nowak K. (2014) Reporting Race and Ethnicity in Genetics Research: Do Journal Recommendations or Resources Matter?. Sci Eng Ethics. 2014 Nov 19 (ahead of print).


  1. Ortega VE, Meyers DA. (2014) Pharmacogenetics: implications of race and ethnicity on defining genetic profiles for personalized medicine. J Allergy Clin Immunol.  Jan;133(1):16-26


  1. Ali-Khan SEKrakowski TTahir RDaar AS. (2011) The use of race, ethnicity and ancestry in human genetic research. HUGO J.  Dec;5(1-4):47-63.


  1. Guidance for Industry: Collection of Race and Ethnicity Data in Clinical Trials." U.S. Food and Drug Administration. September 2005.


  1. Race, Ethnicity, and Genetics Working Group (2005) The use of racial, ethnic, and ancestral categories in human genetics research. Am J Hum Genet 77(4):519–532.


  1. Note for Guidance on Ethnic Factors in the acceptability of Foreign Clinical Data, (CPMP/ICH/289/95), September 1998. ICH Topic E 5 (R1), Step 5.


  1. Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity. Federal Register Notice October 30, 1997.


  1. I.F. Haney Lopez. The Social Construction of Race (chapter 17) in Critical Race Theory. The cutting edge.


  1. Taylor ALZiesche SYancy CCarson PD'Agostino R JrFerdinand KTaylor MAdams KSabolinski MWorcel MCohn JNAfrican-American Heart Failure Trial Investigators. Combination of isosorbide dinitrate and hydralazine in blacks with heart failure (2004) N Engl J Med 351(20):2049-57


Tags: clinical research

Risky Business: Pharmacovigilance and the Role of the NEW PSUR aka PBRER

Posted on Wed, Jan 28, 2015 @ 10:18 PM

By Sandra Drake, MMS Holdings

Pharamacovigilance (PV) is the multi-disciplined science of evaluating the risk benefit analysis of a pharmaceutical compound, which involves monitoring, detecting, assessing and preventing adverse effects of pharmaceutical products or medical devices. The ADR, or adverse drug reaction, is defined as any response to a drug which is noxious and unintended.  Data gathered from patients and healthcare providers, as well as the medical literature, play a pivotal role in PV.  Ultimately, PV is concerned with identifying the hazards associated with pharmaceutical products and with minimizing the risk to patients.   PV can be divided into pre-marketing (also known as Drug Safety activities) and post-marketing activities. Pre-marketing clinical development activities are focused around the identification of Serious Adverse Events (SAEs) and the determination of their potential causal association with a therapeutic agent or device. During post-market activities (once a drug or device is approved and marketed), there is dramatic increase in exposure (or use) of the drug in the population, as well as an increased likelihood of off-label use.   Post-market PV includes a variety of activities centered on identification of “signals” and detection methods which strategically evaluate new potential causal associations between adverse events and use of the therapeutic agent, as required for continuing surveillance of ongoing patient risk. Once a drug or device is approved, PV activities continue to monitor the risk-benefit balance and ensure that it remains favorable to promote and protect patient health and safety.

The annual safety report while a drug or device is still in clinical development is called Drug Safety Update Report or DSUR.  The annual report for post-marketing activities for approved products, known as the US Periodic Adverse Drug Experience Report (PADER) or ICH E2C Periodic Safety Update Report (PSUR), has recently received a major face lift and a new name that better reflects its new contents; it’s now referred to as the Periodic Benefit-Risk Evaluation Report or PBRER.  A recent update to the Guideline for Good Pharmacovigilance (GVP) by the EMA discussed the essential requirements and scope of the PSUR (EMA retains PSUR name, but redefines the PSUR content to that of the PBRER).  The new document requires the summarization of the following types of data and evaluation during the reporting period:


New Data included in PBRER1

  • Critical examination of  new emerging information
  • New relevant safety, efficacy and effectiveness information
  • Integrated risk-benefit analysis for all cumulative information
  • Risk-minimization actions and/or planned activities
  • Updates to any PV plans: risk or signal evaluations

In addition to the format and content specifications, the guideline also recommends inclusion of  reference material for consideration including the company core data sheets  (CCDS) or company core safety information (CCSI) as well the following potential sources of information:  non-clinical studies; spontaneous reports; active surveillance systems; investigations of product quality; product usage data and drug utilization information; clinical trials (including research in off label indications or populations); observational studies (including registries); patient support programs; systematic reviews and meta-analysis; marketing authorization holders sponsored websites; published scientific literature or reports from abstracts (including information presented at scientific meetings); unpublished manuscripts; licensing partners, other sponsors or academic institutions and research networks; and competent authorities (worldwide).   All potential data sources need to be exhausted and data aggregated, analyzed and summarized in the PSUR/PBRER.  In addition to including data from new exposed populations, it is also required to include a summary of safety from alternative or excursions from the recommended use, such as sources that would elaborate the overdose, abuse, or misuse of the therapeutic agent or device.

The updated PSUR/PBRER will include new efficacy (medical benefit) data together with the current risk data and the subsequent cumulative benefit-risk analysis; this is a major addition, as the previous PSUR template/guidance did not include efficacy data. The reporting period of PSUR varies; initially the PSUR may be required bi-annually for  new approved drug product, but later, after the risk-benefit profile is established, a longer interim period may be warranted depending on the risk-benefit status as determined by the health authorities.  All post-marketing pharmacovigilance and surveillance activities are important and one may consider the PSUR/PBRER as one important step of the renewable license, since an important outcome of the assessment may determine the future availability and/or delivery of the drug.  



  1. ICH Guideline for good pharmacovigilance practice, GVP, 2013 April. EMA/816292/2011 Rev 1


  1. Hartmann K, Doser AK, Kuhn M, Postmarketing safety information: How useful are

spontaneous reports. Pharmacoepidemiology and Drug Safety 1999;8:S65-S71.

  1. ICH Guideline E2D; Post-approval Safety Data Management: Definitions and Standards

for Expedited Reporting, 3.1.1 Spontaneous Reports.

  1. ICH Guideline E2E; Pharmacovigilance Planning, 2006.  Spontaneous Reports.






The Year in Review: The Most Exciting Advances in Neuroscience in 2014

Posted on Tue, Jan 13, 2015 @ 10:00 AM

By Charlie Taylor, Regulatory Writer and Drug Development Consultant, MMS Holdings

Neuroscience continues to advance rapidly, with tens of thousands of new peer-reviewed articles each year. Several findings stood out to me in 2014 as being relevant for neurology and psychiatry drug discovery and development. This summary scratches the surface…

Clinical Neuroscience Studies:

  • Cognitive deficits in schizophrenia are a huge unmet challenge because existing antipsychotic drugs do little to help impaired memory, planning for the future and “emotional IQ.”  Santarelli and colleagues, in a trial with schizophrenia patients (Umbricht, Keefe et al. 2014), found that RG3487, an alpha-7 specific nicotinic partial agonist, improved thinking deficits (Negative Symptom subscore) in a subset of schizophrenia patients. Treatment effects were large (about 25% of the total negative symptoms score) and dose-related. These results are very encouraging for future treatment of schizophrenia negative symptoms and also may explain why more than 60% of patients with schizophrenia smoke.
  • Alzheimer’s disease is a huge healthcare challenge that, despite intense drug discovery efforts for over 30 years, remains mostly untreatable. Behavioral agitation, mood swings and aggression in Alzheimer’s are a particular burden to health care providers. Physicians have been criticized for overusing major tranquilizers (antipsychotic drugs) to reduce agitation. Although effective, such treatments sedate patients and compromise memory and thinking. In contrast, a new formulation of dextromethorphan plus quinidine (Nuedexta®) reduced agitation in Alzheimer’s patients without sedation. Dextromethorphan is the active ingredient in most over-the-counter cough suppressants and quinidine is an older cardiovascular drug now used in small amounts to greatly reduce dextromethorphan metabolism. In apparent reaction to the clinical trial news, Avanir stock increased in value 3-fold and Otsuka Pharmaceuticals announced plans to acquire Avanir for about $3.5 billion.
  • New drugs intended to prevent the progression of early Alzheimer’s have been the focus of dozens of drug discovery efforts over the past 20 years. A review of anti-amyloid treatments for Alzheimer’s disease (Karran and Hardy 2014) critically discussed results from 6 different drugs that completed randomized placebo-controlled clinical trials. Overall results were disappointing, with no clear indication that any treatment slowed the advancement of disease. This leaves the field in doubt as to whether any future treatment to reduce amyloid aggregation in the brain will ever successfully treat Alzheimer’s disease.


Laboratory Basic Neuroscience Studies:


  • The 2014 BRAIN Initiative promotes collaboration across U.S. agencies, academic laboratories and the private sector to increase understanding of the human brain using techniques such as large electron microscope serial section reconstructions and large-scale computer simulations that are costly and labor intensive. This initiative supports large studies such as the Human Connectome Project and recording methods that monitor hundreds of brain neurons simultaneously. Older techniques such as single-neuron electrical recordings, although very useful, have been likened to analyzing a football game by watching the responses of a single fan in the stands. Additional steps forward in understanding brain function will require progressively bigger advances in technology.
  • A review of genes that contribute to psychiatric disease (Gratten, Wray et al. 2014) showed that  psychiatric and neurological disorders usually result from many different genes that each contribute a very small risk for disease. For example, over 125 known genes contribute to schizophrenia, 20 contribute to Alzheimer’s disease and 8 to bipolar disorder. No genetic risk factors are known for ADHD, anorexia, major depressive disorder or obsessive-compulsive disorder. In schizophrenia, no single gene contributes more than 4% of the overall risk for disease. In contrast, overall genetic heritability of schizophrenia, major depression and autism are each between 40% and 80% (judged by family tree analysis). Therefore, although heritability of psychiatric disease is relatively high, identification of single genes seems unlikely to uncover future treatments that are useful for the entire disease population. The remaining non-genetic risk of disease is presumed to be caused by environmental factors and de novo mutations and gene duplications that are not inherited. In summary, the hope that sequencing the human genome would lead directly to useful medical treatments for common psychiatric diseases seems unrealistic.
  • John O’Keefe, May-Brit Moser and Edvard Moser shared a 2014 Nobel Prize for their work to describe how neurons in the hippocampus and entorhinal cortex determine position and enable spatial navigation. They discovered “place cells” and “grid cells” that react to visual cues when an animal occupies one spot or moves through 3-D space. This system has been called the brain’s GPS and it also is used to organize memory of sequences of events. Their approach is one of the first to yield new understandings on the mechanisms of abstract thinking that is not directly linked to sensory or motor function.
  • New results helped understand how the brain manipulates the “good or bad” associated with memories. Tonegawa and colleagues at MIT (Redondo, Kim et al. 2014) manipulated a set of neurons in mouse hippocampus and changed the “polarity” of memories from bad (a specific place where mice received foot shocks) to good (a place where mice were exposed to sexually receptive female mice) and vice versa. This was done by selectively activating neurons in the dorsal dentate gyrus of the hippocampus with light pulses that activated neurons while mice were exposed to either positive or negative reinforcement. These results give hope that specific future treatments for post-traumatic stress syndrome and other negative memories might be discovered in humans.
  • During 2014, evidence continues to emerge that sleep disturbances may cause memory problems seen with Alzheimer’s, aging, depression and schizophrenia…And sleep disturbance may contribute directly to the progression of Alzheimer’s disease. These findings suggest that drug treatments or lifestyle changes that specifically improve sleep quality (for example, increasing phase 4 or deep sleep and reducing sleep disruptions) may improve cognitive deficits and also prevent Alzheimer’s disease progression.

o   Artificially increasing a natural protein in the hippocampus of mice (phosphorylated eukaryotic translation initiation factor 43 binding protein or 4EBP2) helps retain spatial memories that otherwise degrade with sleep deprivation. This indicates that the mammalian target of rapamycin (mTOR) biochemical pathway is compromised with sleep loss, causing memory deficits (Choi, Davis et al. 2014). The mTOR pathway is activated by many different stimuli that each cause new transcription of DNA and synthesis of new proteins inside of neurons.

o   Sleep disturbance also may contribute directly to disease progression in Alzheimer’s disease. Maiken Nedergaard and colleagues at University of Rochester (Xie, Kang et al. 2013) found that fluid flow within the brain increases up to 10-fold with slow-wave sleep compared to waking in rats. They propose that metabolic byproducts and amyloid protein accumulate in brain with lost sleep, and this contributes directly to Alzheimer’s disease. In a follow-up paper (Kress, Iliff et al. 2014) they found that brain tissue clearance of amyloid protein is impaired 40% in aged mice compared to young mice. They also showed that the glial ion channel aquaporin-1 is lost in astrocytes of aged mice. The same group (Iliff, Chen et al. 2014) found that head trauma causes glial scars that substantially reduce brain fluid flow. It has long been known that head trauma is a major risk factor for subsequent Alzheimer’s, and this is the first study to directly suggest why.

  • (El-Boustani and Sur 2014) used electrophysiology of single brain neurons and calcium imaging of a large network of neurons together with genetically-engineered light stimulation (optogenetics) to study inhibitory GABA interneurons. The GABA neurons either caused subtractive or divisive inhibition at their target cells, depending on the timing of inhibition with relation to excitation in the neurons they innervate. This finding will provide a better overall understanding and better modelling of neuronal networks in neocortex.

Overall, these advances in neuroscience underscore a couple of conclusions. Firstly, many approaches to treating Alzheimer’s and other serious mental disorders, although based on solid logic and promising results in animal models, have failed miserably in clinical trials. This, unfortunately, is a common fact of life for drug discovery. Secondly, promising findings often emerge from approaches that at first seem trivial (such as improving sleep quality in schizophrenia or Alzheimer’s disease). Undoubtedly, as progress continues toward understanding the brain and mind, future therapies for common disabling diseases such as depression, schizophrenia and Alzheimer’s will emerge that are much different from those in use today. 


1. Choi JH, Davis EJ, Havekes R and Abel T (2014). Restoration of phosphorylated eukaryotic translation initiation factor 4E binding protein 2 (4EBP2) in the hippocampus rescues memory impairment due to sleep deprivation Soc. Neurosci. Abstracts 291.07.


2. El-Boustani S and Sur M (2014). Response-dependent dynamics of cell-specific inhibition in cortical networks in vivo. Nat Commun 5: 5689.


3. Gratten J, Wray NR, Keller MC and Visscher PM (2014). Large-scale genomics unveils the genetic architecture of psychiatric disorders. Nat Neurosci 17(6): 782-790.


4. Iliff JJ, Chen MJ, Plog BA, Zeppenfeld DM, Soltero M, Yang L, Singh I, Deane R and Nedergaard M (2014). Impairment of glymphatic pathway function promotes tau pathology after traumatic brain injury. J Neurosci 34(49): 16180-16193.


5. Karran E and Hardy J (2014). A critique of the drug discovery and phase 3 clinical programs targeting the amyloid hypothesis for Alzheimer disease. Ann Neurol 76(2): 185-205.


6. Kress BT, Iliff JJ, Xia M, Wang M, Wei HS, Zeppenfeld D, Xie L, Kang H, Xu Q, Liew JA, Plog BA, Ding F, Deane R and Nedergaard M (2014). Impairment of paravascular clearance pathways in the aging brain. Ann Neurol 76(6): 845-861.


7. Redondo RL, Kim J, Arons AL, Ramirez S, Liu X and Tonegawa S (2014). Bidirectional switch of the valence associated with a hippocampal contextual memory engram. Nature 513(7518): 426-430.


8. Umbricht D, Keefe RS, Murray S, Lowe DA, Porter R, Garibaldi G and Santarelli L (2014). A randomized, placebo-controlled study investigating the nicotinic alpha7 agonist, RG3487, for cognitive deficits in schizophrenia. Neuropsychopharmacology 39(7): 1568-1577.


9. Xie L, Kang H, Xu Q, Chen MJ, Liao Y, Thiyagarajan M, O'Donnell J, Christensen DJ, Nicholson C, Iliff JJ, Takano T, Deane R and Nedergaard M (2013). Sleep drives metabolite clearance from the adult brain. Science 342(6156): 373-377.




Tags: Year in Review

One Small Secret to a More Successful Briefing Document

Posted on Tue, Sep 23, 2014 @ 08:00 AM

By Ben Kasper, Global Submissions Manager

Many of us who work in the pharmaceutical industry can recall excitedly explaining aspects of our science to friends and family.  Perhaps the discussion took the form of a conversation of food effect around the thanksgiving table, or a detailed breakdown of a mechanism of action to a curious high school biology student.  The questions that arise from discussions are often illuminating—allowing us to see science in a different way and helping us explain it more convincingly in the future.  While discussions with the FDA frequently invoke more anxiety than dinner table conversations, they share one important similarity:  both represent unique opportunities to convey the meaning, challenges, and direction of our work in a broader context, and to a broader audience, than we would normally do within our development teams. The FDA is also key to accepting our conclusions about the science for an effective end-game.


In any communication, proper emphasis on structure and context adds value and is opportunity realized.  When communicating with the FDA, our most powerful tool to build structure and provide context is the briefing document.  The FDA suggests inclusion of 10 types of critical information in the briefing document (Table 1).  Striking the right balance of detail between the related sections is a challenge.  Conceptually, it helps to think of the briefing document as having two major sections:  the background (items 1-9) and the supportive data section (item 10).  The background represents the ability to concisely tell a story that is complete in itself, while the supportive data is the base that allows deeper questions to be addressed quickly and with authority. 

A strong background and thorough supportive data will result in a good briefing document, but not a great one.  To achieve a great briefing document the background sections must also anticipate the reviewer’s questions and direct them to the supportive data in order to answer these questions at the appropriate level of detail.  Here, it helps to have a strong, knowledgeable reviewer outside of the development team to conduct a high level global review.  This reviewer should have sufficient scientific experience and background to comment on the structure and context of the briefing document with a perspective similar to that of an FDA reviewer. It provides a non-passionate objective review. Often, when we are close to the development of a compound, we do not see some of the issues as being questionable even though the science is thorough. A fresh pair of eyes provided by such a review tends to focus on areas of inconsistency, ambiguity, and excessive jargon that accrue in multiple author/draft cycle documents.  They will generally find instances where long searches in the supportive data are required to address questions arising in the background sections.  Eliminating these instances saves time for both the sponsor and the FDA, allowing meetings to focus on content rather than clarification.  Just as an unanticipated question from a friend or family member can lead to new ways of seeing our work, an unanticipated question from an outside reviewer reveal more effectives ways of conveying our work to the agency.

When meeting with the agency, you cannot choose your review team.  One way to prepare for this is to have a trusted reviewer not directly involved in the project to review your briefing document.  With our strong emphasis on science, the medical writers and strategists at MMS Holdings can provide a level of support similar to a trusted outside reviewer, leading to unexpected insights and improved preparation for your most important FDA interactions.