Abstract
Assessment of MRD in APL by Q-PCR analysis for the presence or absence of PML-RARα mRNA after the completion of consolidation therapy in treatment regimens employing all-trans retinoic acid (ATRA) and chemotherapy (CT) has provided a reasonably accurate but not infallible prognostic test for disease recurrence (DR). This assessment has been made at detection sensitivity levels up to 1 in 104, which is considered to have optimal predictive value, since non-quantified lower levels of PML-RARα mRNA may not be relevant to clinical outcome. In the current study, we compared the results of Q-PCR (touchdown PCR for 40 cycles followed by post-PCR blot hybridization with a RARα chemiluminescent probe) to results from RQ-PCR in 15 patients on a Phase II clinical trial designed to eliminate CT (low-risk/LR patients, WBC <10K) or minimize CT (high-risk/HR patients, WBC >10K), using combined ATRA and arsenic trioxide therapy. All patients achieved complete remission (CR) and have been followed for a median of 1.2 years. DR occurred in 0/8 LR, 0/1 unclassified and 2/6 HR patients. At CR, all Q-PCR and RQ-PCR assays were positive. For RQ-PCR, the normalized quotient (NQ) value of PML-RARα, i.e., the PML-RARα copy number divided by the housekeeping gene glyceraldehyde-3′-phosphate dehydrogenase (GAPDH) copy number, ranged widely from 1.5 x 10−6 to 7.0 x 10−3. Post-CR follow-up data were:
Patient Group . | Number . | Q-PCR+ . | RQ-PCR+ . | NQ Range . |
---|---|---|---|---|
LR/Unclassified | 9 | 0/24 | 16/24 | 9 x 10−9 – 1 x 10−7 |
HR/no DR | 4 | 0/23 | 8/23 | 2 x 10−9 – 3 x 10−7 |
HR/DR | 2 | 1/7 | 5/7 | 1 x 10−8 – 5 x 10−5 |
Patient Group . | Number . | Q-PCR+ . | RQ-PCR+ . | NQ Range . |
---|---|---|---|---|
LR/Unclassified | 9 | 0/24 | 16/24 | 9 x 10−9 – 1 x 10−7 |
HR/no DR | 4 | 0/23 | 8/23 | 2 x 10−9 – 3 x 10−7 |
HR/DR | 2 | 1/7 | 5/7 | 1 x 10−8 – 5 x 10−5 |
For the 2 HR/DR patients, the last pre-DR results were, respectively: Q-PCR, positive and negative; RQ-PCR, NQ = 5 x 10−5 and NQ = 2.3 x 10−7. In overall comparisons to RQ-PCR results, all Q-PCR assays were positive for NQ >10−5, 3/4 were positive in the NQ 10−6 to 10−5 range, and all were negative for NQ <10−6. Additionally, dilution-reconstruction experiments confirmed the stochastic nature of PCR assays near the detection limit and indicated that RQ-PCR was about 10-fold more sensitive than Q-PCR. Although larger studies are needed, our results suggest that HR patients who are Q-PCR-negative but RQ-PCR-positive in the >10−7 – <10−5 range at a critical post-CR checkpoint may be at increased risk of DR and, more generally, they suggest that the results of RQ-PCR testing at a critical checkpoint can be used in combination with clinical risk assessment to stratify the intensity of subsequent MRD monitoring in individual patients.
Author notes
Corresponding author
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal