Recently, I have come across a 2-armed clinical trial problem which attempts to find a "Disease Control Rate" (DCR) for patients in the treatement and placebo arms.
On the Internet I found DCR to be defined as follows:
"Disease Control Rate is defined as the percentage of patients with advanced or metastatic cancer who have achieved complete response, partial response and stable disease to a therapeutic intervention in clinical trials of anticancer agents".
Although, I am completely fine with the definition, I am having problems on how to interpret the DCR mathematically.
Say, I have two groups of cancer patients, one receiving treatement and one placebo. Say, intially during treatment, they have abnormal CA-125 levels.
So, should my DCR be the percentage of patients who experienced stable CA-125?
Is there any time limit which I should be considering for acquring this complete or partial response ?
Any article or book which describes the calculation of real time DCR will be very helpful.
Please help me with this.
Best Answer
This is less a question of statistics than it is about outcomes in cancer trials.
A Bit of Background
Outcomes in cancer have evolved over time. In the 1960's the definition of a "response" was a decrease in tumour mass, no increase in an individual tumour's size and no appearance of new lesions (see, for example, [1]). In 1979, the WHO established the bidimensional measurements of tumours as a standard [2] with partial response (PR) arbitrarily defined as decrease of at least 50% in tumour size while tumour progression was defined an increase of at least 25% in size or new lesions. Then, in 1992, the Southwest Oncology Group expanded this idea from an area criterion to a volumetric one [3]. In 2000, tumour response was standardised through Response Evaluation Criteria in Solid Tumors (RECIST), which recommended unidimensional tumour measurement by its longest diameter as a reproducible standard. The definition of PR was a 30% or greater decrease in the sum of the measurable lesions and "progressive disease" was defined a 20% or greater increase or appearance of new lesions [4].
These ways of measuring tumour response all depend on size, which is not ideal is certain circumstances. Thus, the idea of the "disease control rate" was introduced by the Southwest Oncology Group in 2008 [5]. The DCR is a composite measure that incorporates the ideas of a complete response (CR) and PR. Thus, the DCR is a measure of tumour non-progression.
Since its introduction, the idea has been adopted in many areas of cancer, each employing criteria that define this concept of non-progression. The standard inclusion still applies, though. DCR = CR + PR + some measure of stable disease.
Your Questions
If you have evidence to suggest that CA-125 is a marker of disease progression, then, yes, patients who show stable CA-125 will meet the definition of having achieved DCR. However, this is only partially complete, since DCR also includes patients who experienced complete remission and those who experienced partial remission. You cannot only include those with stable CA-125.
The answer to this question is specific to the type of cancer you are studying, including important considerations such as its clinical stage. There is no one single answer for all cancers.
DCR is not calcualted in "real time". Nevertheless, the calculation is quite simple. Let's take this recent article by Ding and colleagues as an example [6]. First, they defined the components of DCR.
Then, the DCR is simply CR + PR + SD or N (the total study population) - PD.
This kind of operationalisation is common.
Good luck.
References