Chemistry: molecular biology and microbiology – Measuring or testing process involving enzymes or... – Involving nucleic acid
Reexamination Certificate
1996-11-29
2001-08-07
Myers, Carla J. (Department: 1655)
Chemistry: molecular biology and microbiology
Measuring or testing process involving enzymes or...
Involving nucleic acid
C435S005000, C435S091200, C435S007100, C435S007900, C435S007950
Reexamination Certificate
active
06270963
ABSTRACT:
BACKGROUND OF THE INVENTION
This application relates to the testing of DNA from a patient sample for mutations, and more particularly to testing for mutations associated with cancer or other diseases for use in diagnosis and targeted screening.
It is becoming increasingly clear that many diseases are caused by genetic mutations. In some cases, these mutations are inherited. In others the mutations are acquired during the lifetime of the individual, for example as a result of exposure to radiation or carcinogenic chemicals. Early diagnosis and optimal treatment of diseases resulting from such mutations will often depend on the ability to detect the mutation, and in some cases to detect the specific nature of the mutation. To this end, various methods have been developed for testing for genetic mutations.
One class of tests makes use of immunodiagnostic techniques such as ELISA to detect the presence or absence of a protein product of the diagnostically important gene. Such tests make use of an antibody which selectively binds to either the normal protein product of the gene or the protein product of the mutated gene, and detect the presence or absence of binding. These tests are generally relatively low in cost. They suffer, however, from low clinical accuracy because they produce many false negative results. This has led workers in the field to question the value of immunodiagnostic tests for diagnosis of or screening for genetic diseases. See Beebe et al.,
J. Clin. Microbiol.
31: 2535-7 (1993); Warren, et al.,
J. Clin. Microbiol.
31: 1663-6 (1993); Roberts et al.,
The Lancet
336: 1523 (1990); De Cresce et al.,
Medical Laboratory Observer
25: 28 (1993); Einstein et al.,
New Engl. J. Med.
322: 178-183 (1990); Hall P. A.,
J Pathology
172: 1-4 (1994).
A second class of tests-for identifying gene mutations in patient samples makes use of nucleic acid probes which specifically hybridize with the portion of the gene containing the mutation site. Probe-based tests have high accuracy (few false negatives) and specificity (few false positives) for the specific mutation. A drawback to probe-based tests however, is this very specificity which requires a priori detailed knowledge of the mutation being tested for, requires a unique set of reagents for each mutation, and may result in the failure to detect new types of mutations. Because of these drawbacks, this class of tests has also been criticized by some as being inadequate for meeting the diagnostic and screening challenges of the future, Ewanowich et al.,
J. Clin. Microbiol.
31: 1715-25 (1993); Hatcher et al.,
Prenat. Diagn.
13: 171-7 (1993); Bull et al.,
The Lancet
340: 1360 (1992).
A third class of tests obtains the full sequence of the DNA for a particular gene recovered from the sample. Erickson, D.,
Scientific American
267: 116 (1992). Rather than infer a diagnosis from indirect probe or protein tests, these tests read the DNA sequence of the gene of interest base by base. This method, which is known as Sequence-Based Diagnosis or SBD has the advantage of near 100% accuracy and 100 specificity. The disadvantage of this method, however, is the cost (approximately $1.00 per base) which effectively renders the method unavailable for screening applications, and even for many diagnostic applications.
As increasing numbers of tests become available for various conditions, diseases and predispositions associated with genetic mutations, the challenge which will have to be faced is the effective utilization of these tests to provide accurate, specific and yet cost effective diagnosis and targeted screening. Dunn et al.,
J. Cell. Biochem
Vol. Supp. 18C February 1994, Page 199 discloses the use of a series of tests to evaluate a sample for mutations in the RB1 gene. No general utility for the test format is disclosed, however, nor is there any teaching of a methodology for means for optimizing the choice of tests for any given mutation.
It is an object of the present invention to provide a method for the establishment of testing “algorithms” which provide the optimum balance of available test procedures for use in connection with any particular genetic mutation.
SUMMARY OF THE INVENTION
This and other objects of the invention are achieved by establishing a hierarchy of at least two different assay techniques which is utilized to assess the presence or absence of a mutation in a gene of interest. Qualitatively, the first assay in the hierarchy is selected to provide a highly specific test for the existence of the disease-associated mutation, although the sensitivity of the test need not be high. The final assay in the hierarchy is selected to provide a highly sensitive and highly specific test for the existence of the disease associated mutation. Intermediate tests of progressively greater sensitivity may also be included in the hierarchy. The selection of an optimum or near-optimum diagnostic algorithm, however, must include a consideration of the cost of each test as well as the sensitivity and specificity of the tests. Furthermore, where more than one or two tests are available to choose from, the selection of the correct set of tests may substantially reduce the cost of the overall testing procedure without loss of performance, while the selection of a different set of tests may fail to provide the same degree of savings, notwithstanding the fact that both sets of tests provide a hierarchical increase in cost and sensitivity.
The present invention provides a mechanism for evaluating test protocols to define a diagnostic algorithm made up of known testing procedures which optimizes overall performance while minimizing costs. In accordance with the invention, the first step in the process is calibration of the available tests for a given genetic mutation to establish a relative sensitivity, specificity and cost for each available test. Each possible combination of tests up to a pre-defined maximum number of tests is then evaluated to determine the overall sensitivity, overall specificity, the predictive value of a positive test and the predictive value of a negative test. For those combinations where these values all exceed a threshold level of reliability, the expected cost of the test, E[C
A
], is calculated using the formula
E
[
C
A
]
=
∑
r
=
1
n
ρ
A
,
r
∑
j
=
1
r
C
(
j
)
where &rgr;
A,r
is the marginal probability that the r-th test will have to be performed to achieve the result and C
(j)
is the cost of the jth test. For any given set of available tests, the preferred diagnostic algorithm or hierarchy is the test which yields the lowest expected cost.
Once the hierarchy has been selected for a given mutation-associated disease, the patient sample is analyzed using each successive assay in the hierarchy. If the result of the first assay is negative for the presence of a disease-associated mutation, then the next assay in the hierarchy is performed. If the result is positive, subsequent assays in the hierarchy may or may not be performed depending on the specificity of the first assay as defined within the diagnostic algorithm. This process is repeated until the final assay has been performed on all samples which gave negative or ambiguous results when tested by all less-sensitive assays in the hierarchy.
A major advantage of the claimed method as described above is the ability to dramatically reduce the per-sample cost of targeted genetic screening and diagnosis. By utilizing methods of progressively greater sensitivity and cost only when the increased accuracy is actually needed, and by selecting a diagnostic algorithm that takes into account the inherent strengths and weaknesses of available tests, the reliability which was thought to be available only by extremely costly method of sequence-based diagnosis can be achieved at average per-patient costs that are a fraction of the cost normally associated with performing a sequence-based diagnosis.
REFERENCES:
patent: 4016043 (1977-04-01), Schuurs et al.
patent: 4172124 (1979-10-01), Koprowski
patent: 4474893 (1984-10
Capatos Denis
Dunn James M.
Matthews David E.
Stevens John K.
Myers Carla J.
Oppedahl & Larson LLP
Visible Genetics Inc.
LandOfFree
Method for testing for mutations in DNA from a patient sample does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for testing for mutations in DNA from a patient sample, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for testing for mutations in DNA from a patient sample will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2551153