Data Management

PHASTAR adds Data Management

By mutual agreement, PHASTAR are pleased to announce the acquisition of Exp-e-Data, an established provider of clinical data management services, to complement our current set of statistics, programming and medical writing offerings. PHASTAR and Exp-e-Data have successfully partnered on many projects for a number of years. Together, we can enhance the services to our customers due to the complementary nature of our companies, building on our shared philosophy, which is to ensure that the work we produce is of optimal quality: every project is supervised on methodology and utilises unique internal processes designed to ensure the highest standards.

The strong working relationships and team bonds forged between both companies over the years have led us to joining forces as a single company so we can grow in a seamless way whilst continuing to deliver a quality service that is ahead of our competitors.

Continue Reading

ASCO 2016

Highlights from this year's conference

The American Society of Clinical Oncology (ASCO) held its annual mega-conference in Chicago, IL this year. With more than 30,000 participants it is the largest international oncology conference. Every year the ASCO meeting takes the measure of the development in cancer research. This year was no exception. The volume of results was truly staggering. There were several important updates that have bearing on the direction of oncology therapeutics.  

Last year's revolutionary advance "Immunotherapy" had many important updates this year. Many of the presentations on Immunotherapy focused on combining Immunotherapies with other cancer treatments. Success of this approach varied but there were some successes particularly in inhibiting tumor growth. Several important updates to which tumors that are susceptible to an immunotherapeutic approach were presented. Genetic data from a range of tumor types pointed to bladder and bowel cancer as good targets although Dr Jedd Wolchock, from the Memorial Sloan Kettering Cancer Center in New York, cautioned during a high-profile immunotherapy session that the approach “is going to take some tinkering before we know what will work for each patient”.

Continue Reading

ASCO16

PHASTAR attending ASCO 2016 meeting

PHASTAR are attending the ASCO annual meeting, which brings together 30,000 oncology professionals from around the world. Educational sessions feature world-renowned faculty discussing state-of-the-art treatment modalities, new therapies, and ongoing controversies in the field.

Clinical studies in oncology have a variety of statistical issues that need to be considered in their design and analysis:

Continue Reading

SAS Art 2015 - The Results

SAS Art Competition 2015 - The Results

Once again, we were amazed by the diversity of artistic talent there is amongst pharmaceutical statisticians and SAS programmers. We had planned to offer one prize, but we have selected a winner and a runner up, as both were extremely close in the final votes.

The Winner

phastarart2015andrewndikom

This piece, inspired by Van Gogh's painting "Field with Poppies" (June 1890), by Andrew Ndikom was voted the winning entry. Andrew describes the work in the following way:

This picture is inspired by the Van Gogh painting Field with Poppies (June 1890). It is constructed from 15000 dots using the pointillism technique. Van Gogh painted several versions of this painting, each one slightly different from the next. The program I have written is able to replicate this approach and is effectively able to produce an infinite number of different versions of the same image. This is achieved through random number generation, with the result that each iteration of the program creates a painting that is unique in terms of its colour, content and composition. Other versions can be viewed in this link.

One of Van Gogh's original versions is shown to the right.

Vincent van Gogh - Poppy field2

Programmatically this is done by creating an annotation dataset, every observation of which determines the properties of an individual point in the painting. The variables of the dataset control the position, colour, transparency and size of each dot and these characteristics are determined by a series of random variables. The cumulative effect of this is that individual images can differ greatly, and that the shade of the field, the shape of the trees, the strength of the sun or the placement of the corn will be different each time.

The sky for example, is built from 6000 points of eight different colours. The colour of each point is selected by generating a random variable between 0 and 1 based on the uniform distribution (using the RANUNI function), if this number falls within a given range then the point is assigned a particular colour. For example if the random variable takes a value between 0.5 and 0.6 then the point generated by this observation will be colour CXAF95A6 (the light colour from the Phastar logo). The position of each point in the sky is similarly determined from a different pair of random variable whose values are re-scaled so that the sky covers the whole image horizontally but only the upper portion vertically.

Continue Reading

SAS Winner 2015

 Festive Prize
 

The winning festive entry was received from Laura Culley a new masters graduate statistician at PHASTAR - this is being used as the design for our PHASTAR 2015 Christmas cards.

Christmas SAS Art - Laura Culley 

Code Here