Starting treatment before the genetics return in suspected acute promyelocytic leukaemia
A 38-year-old man presents with 2 weeks of gum bleeding and bruising. WBC 2.1 x 10^9/L, Hb 78 g/L, platelets 18 x 10^9/L, fibrinogen 0.9 g/L, D-dimer >20 mg/L FEU, PT 19 s, aPTT 34 s, creatinine 118 micromol/L. Marrow aspirate: 25% atypical promyelocytes, many with multiple Auer rods. Flow cytometry shows CD34-negative, HLA-DR-negative, CD33-positive, CD13-positive, brightly MPO-positive cells. PML::RARA RT-PCR has been sent with a 48-72 hour turnaround. The haematology fellow asks whether to start ATRA tonight and whether to insert a tunnelled central line for chemotherapy. The patient is afebrile with no prior bleeding history.
1. Should ATRA be started before the molecular result? What findings would make you doubt the diagnosis?
Read the worked answer
Yes. Morphology (hypergranular promyelocytes, multiple Auer rods) plus the CD34-negative, HLA-DR-negative, bright MPO phenotype is sufficiently typical of APL that ATRA should start tonight, without waiting for PML::RARA. The reasoning is asymmetry of risk: untreated APL kills through haemorrhage, often intracranially, within days, and ATRA is comparatively non-toxic apart from differentiation syndrome. The 48-72 hour wait is not a neutral interval. A rapid PML immunofluorescence or FISH can shorten the wait but must not gate treatment.
Findings that would prompt doubt: CD34 or HLA-DR positivity, absence of MPO or of granules, a monocytic phenotype, or a marrow that looks like NPM1-mutated AML with cup-like nuclei. Those pictures still deserve urgent haematology input but change the therapeutic plan. Conversely, if the PCR returns negative in a patient with classic morphology, do not simply stop ATRA: atypical breakpoints, poor RNA quality and variant translocations all cause false negatives, so request FISH or RNA-based sequencing and repeat the marrow.
2. How would you manage the coagulopathy, and what do you advise about the central line and other procedures?
Read the worked answer
Treat the coagulopathy aggressively and expect ATRA to improve it over several days, so bridge rather than wait. Transfuse to keep platelets above roughly 30-50 x 10^9/L (higher, 50, if there is bleeding or an invasive procedure planned) and fibrinogen above 1.5 g/L with cryoprecipitate or fibrinogen concentrate; check fibrinogen at least daily and after every product. Do not chase the PT/INR as a target and do not give prophylactic heparin for the D-dimer.
Defer the tunnelled central line until platelets and fibrinogen are adequate and the coagulopathy is clearly improving; use a peripherally inserted or temporary line in the interim if access is needed. Avoid intramuscular injections, arterial punctures where avoidable, and any elective procedure. Do not perform leukapheresis: it is not indicated at this WBC and adds risk.
Anticipate differentiation syndrome (days 2-21: fever, weight gain, dyspnoea, infiltrates, pleural or pericardial effusion, renal impairment). Monitor daily fluid balance and oxygen saturation; treat with dexamethasone 10 mg twice daily and hold ATRA temporarily if severe. Hyperleukocytosis is managed with hydroxyurea rather than apheresis.
3. PML::RARA is negative on two occasions but the marrow morphology is unchanged and the coagulopathy persists. How do you proceed?
Read the worked answer
This is the point where the diagnosis rather than the treatment becomes the question. First, verify the specimen: was the PCR done on an adequate cellular sample, was the RNA intact, and was a validated assay with known breakpoint coverage used? Repeat on a fresh aspirate or trephine with FISH for t(15;17) and, if available, RNA sequencing or a broad fusion panel, because cryptic or atypical breakpoints are a recognised cause of false-negative RT-PCR.
Consider the variant APL syndromes. t(11;17) ZBTB16::RARA produces a promyelocytic leukaemia that is ATRA-resistant, with more numerous and finer granules and frequent CD56 expression; it responds poorly to ATRA and to arsenic, and needs different therapy. t(5;17) NPM1::RARA is ATRA-sensitive. Also reconsider AML with NPM1 mutation, which can mimic APL morphologically but lacks the coagulopathy pattern and responds differently.
Practically: continue supportive transfusion while the workup is expedited, keep ATRA only if morphology remains classic and the alternative is worse, and make the decision jointly with a reference laboratory and the patient's haematologist. Do not let a negative PCR alone convert a classic APL into standard AML chemotherapy without an alternative explanation.