Wednesday, February 25, 2015

The Role of Study Data in a Growing Medical Device Market


Medical Device trials are sometimes considered less intricate than drug development studies, however medical device studies have complexities that span across regulatory challenges, product types and unique statistical analyses and study designs.
AdvaMed, an association of medical device makers, released a report saying that the medical device industry has the potential to reach $30 billion USD in the next ten years "if proper regulatory framework is provided to the sector".
This being said, challenges are present for both device developers and consumers. Consumers are after an affordable product that is available as early as possible, while device makers are anxiously seeking funding in the initial stages and then looking to streamline the clinical trials process to get a product to market quickly and within budget. In the middle are the regulatory authorities who throw up roadblocks when it comes to patient safety and device efficacy.
As of 2014, the EMA is requiring more sophisticated clinical trials by eliminating the equivalence principal as an acceptable market-ready justification and proposing more post-market follow up. These reforms mirror the process already in place by the FDA, however the FDA has recently reached a decision on data systems and will not regulate devices that merely "display, store or transfer information". The decision is the FDA's attempt to not meddle in software and IT and promote more innovation in the sector.

Therefore, how important is a  data strategy in this growing sector? Medical Device companies need to think about strategy from conception to market - from preparing due diligence ready datasets for funders to post-market safety surveillance for regulatory authorities.
The medical device market is growing because of market entry from new startups looking to introduce new, innovative products to the market. However, these startups often need to secure funding for further development. Preparing your datasets in a due diligence manner is an important first step and companies should consider the following areas:

How to outsource your study data:
Centralizing study data through a specialized company: centralizing data in the early stages of device development promises common assessment methods, uniform traceability of data and the centralization of study metrics

Functional Service Provision: relying on a specialized clinical data company to provide a scalable team of resources at hourly or daily rates for a determined period of time.

Preparing your clinical data:
Implementing CDISC standards helps both traceability and cross analysis of datasets. There must be clear traceablity from analysis results, to analysis datasets and to SDTM datasets. The FDA announced in 2014 that it is establishing guidelines on the use of CDISC for electronic submissions of study data in applications and it is more likely that guidelines requiring CDISC standard conformity is in the near future.

Implementing Technology:
Technologies such as ePRO and EDC can help keep costs under control, especially if there are low-cost options. Implementing an EDC solution into your medical device trial can provide immediate feedback, more accessible trial information, and higher data quality. It can facilitate an adaptive approach, or in the case of a trial that is showing a negative trend, it can allow for early termination, thereby reducing the risk to patients and reducing the cost of a trial.

Implementing a cloud-based solution improves the quality of data through automated edit and discrepancy checks, faster data cleaning, audit trails and compliance with GCP and regulatory guidelines.

Statistical Consultants for Regulatory Authority Communication
The FDA has made a recent effort to improve communication with medical device markets. Preparing datasets and reports for regulatory submission are crucial steps for medical device companies, and they should have expert consultants to help. Involving an expert biostatistician from the beginning can help with the protocol development and study design.

Since statisticians involved in medical device trials must balance potentially expensive products, long follow up periods, possibly large sample sizes and unique endpoints, sophisticated trial design can help account for these changes. The FDA accepts the use of Bayesian design in medical device trials since it combines data from previous studies and the ongoing study to make changes to the study if necessary.

Consultant statisticians can also facilitate the communication between the device makers and regulatory authorities by interpreting and explaining clinical trial design and results.

CROS NT has extensive experience, both in the U.S. and Europe, in the area of medical device study design and analysis for regulatory submissions. We have worked with multiple startups and late stage device companies including work on novel, Class III devices and consulting on de novo Class II devices. If you need trial design and statistical assistance on an upcoming study, inquire about an initial consultation with one of CROS NT's principal biostatistician.

CROS NT is exhibiting at MedTec Europe 2015 in Stuttgart, Germany this April at Booth #3H55. Be sure to visit us at Europe's premier medical device conference to learn about our expertise in biostatistics and clinical data management and how we can assist your company. 

Tuesday, February 3, 2015

World Cancer Day 2015: The Role of Biostatistics in Oncology Research

The 4th of February is recognized around the world as World Cancer Day, and this year's theme is Not Beyond Us" which focuses on the positive and proactive approach to cancer and the fact that solutions and treatments do exist - solutions that CROS NT is regularly involved in through the analysis and reporting of crucial clinical trial data from oncology studies.

Cancer is among the 10 leading causes of death in the world and accounts for approximately 8.2 million deaths a year in the U.S. Oncology drug development consumes an average of seven years in clinical evaluation and yields a discouraging success rate: only 7% of oncology agents entering Phase I gain marketing approval. 

What makes oncology so unique?

It is not a single characteristic which makes cancer research different, but rather the simultaneous occurrence of: 
  • Many etiologies
  • The importance of disease sub-types and/or genotypes
  • The use of treatment combinations
  • Different modes of action of drugs
  • Regimen modifications during treatment
  • The high impact of the disease on patient life
  • The high costs of treatment
  • Slow recruitment
  • Long timelines to reach clinical endpoints
The Role of Biostatistics in Oncology Research

How can a statistician help drug developers focus on the right dose, the right disease in the right patients as early in the research process as possible?

Early Phase Success

Phase I and II product development lays the foundation for successful Phase III studies. Particularly in oncology, the heterogeneity of the disease can influence the planning of clinical trials. New study designs make it possible to minimize the number of patients in the early stages.

The Statistical Analysis Plan for oncology trials, especially for Phase I trials, is extremely important in terms of determining trial design, sample size, endpoints and determining inclusion/exclusion criteria. The biostatistician should be involved in the beginning of an oncolody study to consult on these critical aspects.  

Good practice designs in oncology can speed up the planning phase; a team with previous oncology experience will enable a reduction of time from the study synopsis to first patient in the study because it defines adequate target criteria, interim analyses and specifices the most efficient statistical method for analysis.

Late Phase Oncology Study Designs

Given the unique characteristics of oncology trials, Adaptive Trial Design can be an ideal statistical methodology for more efficient, cost-effective trials, especially in Phase III development where efficacy should be confirmed. It allows for the selection of the right target population for the drug. This is of special interest in the development of highly specific cancer drugs, which are only effective in a selected patient population, for example monoclonal antibodies. The advantages of adaptive design for oncology research are a reduction in overall time in the development of a drug, fewer patients required and early availability of long-term safety data.
Clinical Endpoint Considerations

Oncology is also unique in that the time to reach clinical endpoints is longer than other areas. The ultimate endpoint for registration of a new drug is still the "Oveall Survival Time". Surrogate endpoints like "response rates" or "time to progression" are often used but mostly as secondary endpoints or in earlier phases of the development process.
The long time period before any conclusion from the study can be drawn is the main reason for the use of adaptive designs in this setting. Adaptive design enable researchers to make early decisions on interim analyses such as sample size adjustment or early termination if a study is doomed for failure.

CROS NT and Oncology Research

Oncology is a particular area of expertise for CROS NT, and we have completed over 100 studies in multiple oncology indications from small Phase I trials to large Phase III and IV multinational trials. Our biostatisticians have worked in numerous indications including breast cancer, prostate cancer, colorectal cancer, CML, AML, lung cancer, lymphoma and leukemia.
CROS in the Community - World Cancer Research Fund

CROS NT is sponsoring the World Cancer Research Fund as its "CROS in the Community" project throughout 2015. The World Cancer Research Fund (WCRF) focuses on preventative cancer solutions by exploring the link between diet, weight, physical activity and cancer, and WCRF operates on an international level. CROS NT is making both a company-wide and individual effort to raise funds and awareness for WCRF this year, some of our initiatives include promoting the cause at industry conferences throughout 2015, as well as individual employees raising awareness by participating in sponsored events.

Thursday, January 29, 2015

ICT Article Feature | Risky Business: Strategic Collaborations Between Sponsors and CROs

In this month's edition of International Clinical Trials (ICT), CROS NT's Global Head of Business Development and Marketing, Chris Hamilton, talks about strategic outsourcing models between Sponsors and CROs. Chris addresses the risks of the strategic partnership approach and how to build an effective relationship and sustainable outsourcing model with CROs.

Strategic collaborations between sponsors and large CROs are now the norm, but trial partners risk being locked in with no room for manoeuvre - which can spell danger in this ever-changing market. Looking to a more flexible, long-term model must become more of an industry priority.

The clinical outsourcing world has always moved in cycles. Drug development companies - particularly the bigger players - shift from building large internal teams, to blending insourced staff with permanent teams, onto outsourcing projects or programmes, passing through functional outsourcing and arriving at strategic partnerships - before starting the whole process again. 

CROs and their customers need to be aware of the risks from the strategic partnership approach. On the CRO side, we have already seen evidence of shareholder dissatisfaction where predicted revenues from a partner do not meet market expectation. CROs cannot rely on only one or two partnerships because this makes them vulnerable; the problem is that, once parties agree multiple partnerships, individual sponsors need to ask: is my relationship still 'strategic'?

To read the full article from International Clinical Trials, click here.

For more information on CROS NT's outsourcing models or to submit an RFP, please visit our website at www.crosnt.com

Thursday, January 15, 2015

CROS NT Biostatisticians Co-Author Moderate Asthma Study Results in The Lancet Medical Journal

CROS NT’s Chairman, Paolo Morelli, and biostatistician, Marco Pannacci, are co-authors on a recent paper published in the The Lancet Respiratory Medicine featuring the results of a moderate asthma clinical trial sponsored by the Italian Medicines Agency (AIFA).

The full article can be found at http://www.thelancet.com/journal/lanres/issue/current in the Lancet’s most recent issue.

Paolo Morelli, given his statistical experience and background in clinical research, provided expert methodological input in the planning of Study Design, Protocol Writing, Sample Size Calculation and reviewing all statistical topics of the study.

Marco Pannacci worked as lead biostatistician on the study, planning and writing the Statistical Analysis Plan and Data Review Report, providing his support and his statistical view during the programming phase and performing statistical reviews of the final results. Marco collaborated with the main study Investigators, further expanding his experience and knowledge in the area of respiratory diseases.

The non-inferiority, randomized, double-blind clinical study screened a total of 1,010 patients between April 2009 and March 2012 with moderate persistent asthma and monitored whether a combination treatment administered “as needed” was as effective as regular treatment using the same maintenance and reliever therapy (SMART).

This study was unique in that “no previous study has investigated whether moderate asthmatics can be just as well controlled by a symptom-driven inhaled corticosteroid and long-acting beta2 agonist combination therapy in the absence of regular maintenance treatment”.

The data for the study were collected by clinical investigators and analyzed by CROS NT statisticians. The statistical analysis consisted of Kaplan-Meier estimates to assess the time to treatment failure and the probability of patients with no treatment failure at 1 year. Cox proportional hazards regression model was also used to analyze time to treatment failure. The proportional hazards assumption was assessed by using both the curves of log(–log[survival]) versus log of survival time and with the correlation of the scaled Schoenfeld residuals on functions of time.

The primary endpoint, which was time to first treatment failure measured after 1 year, was assessed in both the intention-to-treat and per-protocol population.

Conducting Respiratory Trials – a Look at Asthma
Asthma is one of the largest indications being researched in respiratory clinical trials given that around 300 million people suffer from some form of asthma worldwide. There is also a rising demand for prescription devices such as inhalers and nasal sprays for asthma treatment.

CROS NT has over 20 years of experience in conducting respiratory clinical trials and has performed over 100 trials in Phases I-IV resulting in over 50,000 patients managed in indications such as asthma, COPD, Pulmonary Cystic fibrosis, pneumonia and respiratory allergies. Our biostatisticians and data managers are well-experienced in projects involving the integration of electronic devices with EDC systems. Our solution of implementing a Spirometric device with an EDC system to record health measures and diary data has been fully validated and used in respiratory trials.

Tuesday, January 13, 2015

CROS NT Launches New Logo and Website as part of Rebranding Initiative

CROS NT today announced the launch of a new logo and a new website as part of its rebranding initiative to reflect the company’s position as a global, united brand.



The branding initiative comes after a year of restructuring following the acquisition of two biometrics CROs – MDSL International in the United Kingdom and Stat-Tech Services in Chapel Hill, North Carolina. These acquisitions have expanded the CROS NT brand internationally as well as adding a wealth of expert resources in biostatistics, clinical data management and medical writing to its team. As of January 2015, CROS NT is transitioning from the MDSL and Stat-Tech brands to the CROS NT brand name.

CROS NT remains focused on clinical data services – including biostatistics, statistical programming, clinical data management and medical writing – as well as accompanying technology such as ePRO, EDC, IWRS, CTMS and data visualization solutions for the pharmaceutical, biotechnology and medical device sectors.

Chris Hamilton, Global Head of Business Development & Marketing for CROS NT, said, “the rebranding initiative is an important step in the history of CROS NT as it demonstrates the growth in our company over the past few years and marks us as a global player in the CRO market. CROS NT is experiencing great progress due to continued excellent customer service and in high quality clinical data services”.

Andrew MacGarvey, CEO, added, “the new design brings our branding up to date while retaining elements of our original logo to ensure our strong heritage is still represented”.

The new website is live today at www.crosnt.com.

Monday, December 15, 2014

What is the "Ideal" EDC System for a Clinical Data Manager?

Conducting a clinical study requires the collection of a large amount of data that can be entered, checked, confirmed and finally grouped together in a single database. Setting up a central database and electronic collection via the Internet called eCRF (electronic case report forms) meets these needs by combining technology, simplicity, privacy and data security.

According to a study by CenterWatch, if the market continues to adopt EDC and ePRO technologies, paper diaries will no longer exist in 10 years. Should paper diaries still be used to collect subject data in clinical trials? CROS NT's expert Clinical Data Management team discusses this issue and what the ideal EDC system is for data collection.

There are several challenges to using paper CRF:
  • Massive amount of paperwork
  • Tedious for Investigators - repetition, data transfer, consistency
  • Tedious for Monitors
  • Logistical issues
  • All data must be transcribed
  • Data cannot be evaluated in real-time
  • More time and effort needed for data cleaning
  • Extended timelines from LPLV to DB Lock
eCRF significantly reduces the time and costs for certain stages of a clinical study. Direct and indirect costs can be reduced thanks to parallel operations during the use of a centralized database that can be accessed simultaneously by many people using the Internet.

Choosing the Ideal EDC System
Conducting studies in accordance with industry guidelines like ICH and GCP leads to reporting and documentation requirements that are burdensome and complicated. There is no real guidance from regulatory agencies on how to evaluate the many EDC systems available.


The availability of cost-effective, open source or proprietary EDC applications have the potential to improve clinical and research activities. User-friendly and simple interfaces, adherence to industry standard security protocols, customization and low maintenance costs are some of the major benefits.


Commercial EDC systems can be expensive in comparison to open source or proprietary EDC systems. It is important to evaluate and find the functions that best fit data management needs.

What do Data Managers want in an EDC System?

  • An EDC sytsem compliant with regulatory requirements: audit trails, data validation, system integrity
  • A system than enhances communication and sharing with users
  • Flexibility to allow for study specific customizations and dynamic forms
  • Intuitiveness and user-friendly system that is easy to navigate
  • Intuitive to select visits and enter data for end users
  • Easy to retrieve data queries
  • Efficient data cleaning and easy to update
  • Easy for study monitors to perform and mark SDV
  • Easy for Investigators to sign when CRFs are complete
CROS NT can also offer a cloud-based eClinical application which is an integrated platform of EDC/ePRO/IWRS/CTMS. Combining an eClinical platform in the cloud with a centralized approach to clinical data, Sponsors can reduce set-up, training and HelpDesk costs and allow for easy data integration. A cloud-based solution allows for real-time study management access to data from mobile devices.

Monday, December 1, 2014

5 Things to Consider in a Global Data Strategy

Most clinical trials today are being conducted on a global level with vast amounts of data to collect, analyze and report. Trial strategies include collecting data from multiple sources and sites, ideally in real time so Data Managers can manage discrepancies and unusual data patterns and Biostatisticians can analyze incoming data and make critical decisions on trial progress.

Therefore, when implementing a global data strategy, what factors should Sponsors consider in order to ensure efficient project management, timely and quality data and cost-effective measures? CROS NT guides you through some considerations:

1. Which technology solutions are best for managing vast amounts of data?

The obvious solution may be to consider an EDC solution for clinical data management which can organize multi-language, global study data into a single database.. Sponsors can resolve discrepancies faster and reduce time and costs with immediate feedback from patients. Data can be transferred in real time during the study into the eCRF. Query management is accelerated, and any inconsistencies in CRF or ePRO data can be checked in real time.

A cloud-based EDC system can eliminate the need to invest in an IT infrastructure while reducing training efforts. Cloud-based clinical applications also guarantee better data quality and data security through secure data transfer connections.

However, Sponsors should also consider eClinical platforms that include ePRO integration, CTMS and IWRS to manage all aspects of a clinical trial. It is important to note that not all EDC systems and/or eClinical applications fit all studies, so Sponsor should consult their CRO on which system is best.

2. Consider a Centralized Statistical Monitoring Approach

Centralized monitoring has been proposed as a new, more efficient approach to on-site visits. A CSM approach can be useful in detecting faulty equipment errors, negligence or fraud, protocol deviations and unexpected patterns which then identifies the sites that need further investigation. CSM has been suggested as a cheaper and more efficient alternative to on-site monitoring. Combining CSM with EDC and Clinical Data Visualization solutions can make monitoring extremely efficient and cost effective.

3. Should clinical data be centralized?

A centralized biometrics approach promises the Sponsor:
  • One relationship between the biometrics CRO and Sponsor and team continuity
  • Better integration of studies across all phases with common assessment methods and data standards
  • Uniform traceability of data
  • One set of biometrics Standard Operating Procedures
  • Centralization of study metrics and reporting
  • Cost reduction through the re-use of global libraries (savings up to 40%)
4. Building the Best, Most Efficient Project Team


The clinical data team is crucial to success and it’s important to have team continuity throughout global data projects. Sponsors should work closely with CROs to put together the best team of project managers, statisticians, data managers and medical writers. By centralizing biometrics, Sponsors can be ensured of the same biometrics team from start to finish. This is particularly important when it comes to the biostatistician who can provide insight from protocol to reporting through all phases of the development process.

Sponsors can also look to implement the Functional Service Provision model in which teams are assigned to a certain project for a fixed period of time. This allows the Sponsor to cover peaks and troughs in the workload while having a dedicated biometrics team.

5. Making Sense of Clinical Data: Biometrics Consultancy

Many challenges present themselves through the drug and device development process. Sponsors can’t be prepared for all issues that many arise: from trial design and sample size recalculation to regulatory challenges, sometimes Sponsors need help making sense of clinical data on a case-by-case basis. Having an expert biostatistician, data management expert or quality assurance professional available for consultancy can help Sponsors resolve small issues that arise during trials but have the potential to make a big impact.