The cause for the discordant advia centaur xp ca19-9 results is unknown.Siemens healthcare diagnostics is investigating.The ifu states in the limitations section: "warning do not use the advia centaur ca 19-9 assay as a screening test or for diagnosis.Do not predict disease recurrence solely on levels of advia centaur ca 19-9.Normal levels of advia centaur ca 19-9 do not always preclude the presence of disease.Note: do not interpret serum levels of ca 19-9 as absolute evidence of the presence or the absence of malignant disease.Before treatment, patients with confirmed gi carcinoma frequently have levels of ca 19-9 within the range observed in healthy individuals.Additionally, elevated levels of ca 19-9 can be observed in patients with nonmalignant diseases.Measurements of ca 19-9 should always be used in conjunction with other diagnostic procedures, including information from the patient's clinical evaluation.".
|
Siemens filed the initial mdr 1219913-2017-00200 on september 6, 2017.Siemens filed the mdr 1219913-2017-00200 supplemental report 1 on september 20, 2017.09/21/2017 additional information: the customer performed a lot to lot comparison on august 22nd by testing patient samples with the (b)(6) ca 19-9 lot 402 that had previously been tested with the (b)(6) ca19-9 lot 398 and saw a bias.The customer cannot evaluate lots 398 and 402 again because they do not have more of lot 398.The customer stated they stored their samples very well and they do not believe testing on different days explains the bias.The customer is using lot 402 but they have implemented a correction factor they calculated to account for their perceived bias with lot 402.Siemens has informed the customer not to use the correction factor.The control data for lot 398 on (b)(6) 2017 provided by the customer for review shows multiple control replicates recovering > 2 standard deviations (sd) below target and sometime even >3 sd.This control recovery prompted recalibration of lot 398 on (b)(6) 2017.A shift in their system performance on (b)(6) 2017 is suspected.However, quality control was not run with lot 402 on (b)(6) 2017.So, the customer did not know if their system performance shifted and affected lot 402.If the customer had tested quality controls with lot 402 on (b)(6) 2017, it is likely the controls would have failed causing the customer to recalibrate lot 402.If the customer had recalibrated lot 402 on (b)(6) 2017, there may not have been a bias between lots 398 and 402.The data generated with patient samples tested with lots 398 and 402 when lot 402 was released, does indicate lot 402 recovers slightly (~6%) lower than lot 398.The bias seen internally was acceptable for this product.The cause for the discordant (b)(6) ca19-9 results can not be determined.When performing comparisons between lots, it is best practice to test the samples on both lots on the same day.If this is not possible the customer should at least be verifying system performance to make sure their testing is valid.Siemens does not support using a correction factor.The instrument is performing within specifications.No further evaluation of the device is required.
|