HK J Paediatr (new series) 2019;24:203-215

Comparison on Treatment Outcomes on Paediatric AcutePromyelocytic Leukaemia: ICC APL 2001 Versus HKPHOSG

AML 1996 Protocol

WYK CHAN, GKS LAM, JKH CHIU, DTL KU, KKH HO, SY HA

ON BEHALF OF THE HONG KONG PAEDIATRIC HAEMATOLOGY AND ONCOLOGY STUDY GROUP

Abstract Acute promyelocytic leukaemia (APL) is a biologically and clinically distinct variant of acute myeloidleukaemia (AML). Patients often presented as a medical emergency with lethal haemorrhages. Despitehigh early mortality rate, APL has superior event-free survival (EFS) with prompt administration of all-trans retinoic acid (ATRA). This study is a retrospective review comparing outcomes of paediatric APLpatients in Hong Kong treated with the two previously adopted protocols (HKPHOSG AML 1996 versusICC APL 2001) over past 20 years. Total 53 eligible patients were identified, 30 and 23 were treatedwith HKPHOSG AML 1996 and ICC APL 2001 protocol respectively. Five-year overall survival andEFS for HKPHOSG AML 1996 versus ICC APL 2001 protocol were 80% versus 82.6% and 66.7%versus 75.1% with a median follow up period of 193.4 versus 56.7 months. To conclude, local datareveals that ATRA-based therapy demonstrate better outcome than AML-based chemotherapy in treatingpaediatric APL patients.

Key words Child; Drug therapy; Hong Kong; Leukaemia; Survival

Original Article

Department of Paediatrics, Hong Kong Children's Hospital,1 Shing Cheong Road, Kowloon Bay, Kowloon, Hong KongSAR

WYK CHAN MBBS, MPHGKS LAM MBBSDTL KU MBBSSY HA MBBS

Department of Paediatrics, Queen Mary Hospital,102 Pokfulam Road, Pokfulam, Hong Kong SAR

WYK CHAN MBBS, MPHSY HA MBBS

Department of Paediatrics, Prince of Wales Hospital,30-32 Ngan Shing Street, Shatin, N.T., Hong Kong SAR

GKS LAM MBBS

Department of Paediatrics, Queen Elizabeth Hospital,30 Gascoigne Road, Kowloon, Hong Kong SAR

JKH CHIU MBBS

Department of Paediatrics, Tuen Mun Hospital, 23 TsingChung Koon Road, Tuen Mun, N.T., Hong Kong SAR

DTL KU MBBS

Department of Paediatrics, Princess Margaret Hospital,2-10 Princess Margaret Hospital Road, Lai Chi Kok,Kowloon, Hong Kong SAR

KKH HO MBChB

Correspondence to: Dr WYK CHANEmail: wykchan@hku.hk

Received August 8, 2019

Outcomes of Paediatric APL in HK204

Introduction

Acute promyelocytic leukaemia (APL) is a rare formof acute myeloid leukaemia (AML) with only around 2 to4 paediatric cases annually in Hong Kong. APL wasclassified as AML-M3 in the older French-American-British (FAB) classification system and is currentlyclassified as APL with t(15;17)(q24.1;q21.2); PML-RARAin the World Health Organization (WHO) classificationsystem.1 APL is a biologically and clinically distinct variantof AML. Patients often presented as a medical emergencywith lethal pulmonary or cerebrovascular haemorrhage dueto coagulopathies (disseminated intravascular coagulationand primary hyperfibrinolysis) induced by leukaemicpromyelocytes. Despite high rate of early mortality, APLhas superior EFS of above 70% to nearly 100% across allage and risk groups once the differentiation agent all-transretinoic acid (ATRA) is promptly administered whendiagnosis is timely suspected or confirmed.2

Due to rare local incidence of paediatric APL, the HongKong Paediatric Haematology and Oncology Study Group(HKPHOSG) had adopted the HKPHOSG AML 1996protocol based on UK MRC AML 10/12 protocol intreating territory-wide paediatric APL patients in all localpaediatric oncology centres during 1997 to 2008,followed by joining the International Consortium forChildhood (ICC) APL 2001 protocol as one of theparticipating regions in the ICC since 2009 till present.

With the establishment of the Hong Kong Children'sHospital (HKCH) as the single territory-wide paediatriconcology centre in Hong Kong after merging the fiveexisting paediatric oncology units − Queen Mary Hospital(QMH), Prince of Wales Hospital (PWH), Queen ElizabethHospital (QEH), Tuen Mun Hospital (TMH) and PrincessMargaret Hospital (PMH); together with advancements inunderstanding on biology and treatments of paediatric APL,this study serves as a retrospective review on the outcomesof paediatric APL patients in HK treated with the twopreviously adopted protocols (HKPHOSG AML 1996 andICC APL protocol 2001) providing scientific evidencebefore initiation of new treatment protocol.

Methods

All children diagnosed to have APL at age 18 or belowat the time of diagnosis during the period between January1, 1996, and December 31, 2017 in the 5 local paediatriconcology centres were included. Accrual of demographic

data and outcomes were conducted by HKPHOSG dataofficers. Cross-checking of clinical information wasconducted via the Electronic Patient Record (ePR) throughthe Clinical Management System (CMS). Clinical DataAnalysis and Reporting System (CDARS) was alsosearched using the diagnostic code for Acute MyeloidLeukaemia, M3 (ICD-10 205.00) for any missing cases.

Statistical analyses were performed with the StatisticalPackage of Social Sciences version 24.0 (IBM SPSS24.0). Continuous variables were expressed as mean andstandard deviation. Overall survival (OS) was calculatedfrom the date of diagnosis to the date of death of any causeor last follow-up (censored). Event-free survival (EFS) wascalculated from the date of diagnosis to date of relapse,death of any cause, or last follow-up (censored). Theprobabilities of OS and EFS were calculated using theKaplan-Meier estimation and log rank test was used tocompare the survival of different groups. P-values <0.05were deemed statistically significant.

This study complies with the Declaration of Helsinkiand approval from Institutional Review Board of theUniversity of Hong Kong / Hospital Authority Hong KongWest Cluster had been sought (UW 14-890).

Results

Total 53 eligible patients were identified, 30 and 23 wererecruited and treated with HKPHOSG AML 1996 and ICCAPL Study 2001 protocol respectively (see Table 1).Median age of diagnosis was 12.2 years (range 2.4-17.8years). Male-to-female ratio was 1.1 to 1. Two patientsdied before treatment in both treatment arms. Another twopatients died during induction phase chemotherapy in bothtreatment arms, contribute to 6.7% and 8.7% of eligiblepatients in each cohort respectively. Remission wasachieved in the remaining 26 (86.7%) (20 in CR1, 4 in >CR2with 6 deaths) and 19 (82.6%) (15 in CR1, 1 in >CR2 with 4deaths) respectively. There were total 5 (16.7%) and 1(4.3%) patients relapse respectively. The 5-year OS and EFSfor HKPHOSG AML 1996 versus ICC APL Study 2001protocol was 80% vs 82.6% and 66.7% vs 75.1% with a medianfollow up period of 193.4 vs 56.7 months (see Table 2).

Discussion

A. EpidemiologyAPL, FAB classification M3 or M3v, represents

Chan et al 205

approximately 4-8% of paediatric AML1 but the incidencemay be higher in children of Hispanic and Mediterraneanorigin. The median age at presentation is probably similaras that of other AML subtypes, but APL has rarely beenreported in the first year of life. It can arise de novo or betherapy related (t-APL). The characteristics and outcomeof t-APL appear similar to those of de novo APL.3

APL accounts for 5 to 20 percent of AML dependingon geographical location, with higher incidence amongHispanics in Central and South America, Italy, and Spain.Blacks had lower lifetime incidence rates than non-Hispanic Whites, Hispanics, and Asians.4-8 In Hong Kong,four or less cases were diagnosed every year, whichaccounts for 16.7 percent (23 out of 138) of all AML cases

diagnosed in recent 10 years (see Table 1). Various studieshave reported a modest increase in incidence of APL inthe last two decades8,9 but this trend is not observed in ourlocality. As for age distribution, its incidence is rare inthe first decade of life, it then increases during seconddecade, reaches plateau during early adulthood, and thenremains constant until it decreases after the sixth decade.9

For local paediatric cohort, two-thirds of them presentedin second decade of life. There is no gender predilection.APL sometimes occur as a subgroup of therapy-relatedmyeloid neoplasms (t-MDS/t-AML) after cytotoxictherapy (especially topoisomerase-II inhibitors such asetoposide, doxorubicin, and mitoxantrone) or radiationtherapy.10

Table 1 Total number of eligible patients entered into IPC ALL 2001 studyClassified by year

Total number of cases 2009 2010 2011 2012 2013 2014 2015 2016 2017 Total

APL 3 1 4 4 0 1 2 4 4 23AML 18 19 19 15 15 8 16 14 14 138%APL/AML 16.7 5.2 21 26.7 0 12.5 12.5 28.6 28.6 16.7

Classified by hospital

PMH PWH QEH QMH TMH Total

Total number of cases 4 3 5 6 5 23PMH: Princess Margaret Hospital; PWH: Prince of Wales Hospital; QEH: Queen Elizabeth Hospital; QMH: Queen Mary Hospital; TMH: Tuen Mun Hospital

Table 2 Comparison of treatment outcomes (ICC APL Study 2001 vs HKPHOSG AML 1996)

Protocol ICC APL Study 2001 (n=23) HKPHOSG AML 1996 (n=30)

Patients eligible 23 (100%) 30 (100%)Patients died before treatment 2 (8.7%) 2 (6.7%)Remission achieved 19 (82.6%) 26 (86.7%)Death in induction 2 (8.7%) 2 (6.7%)Non-responders 0 0Relapse 1 (4.3%) 5 (16.7%)Survivors CR1 15 (65.2%) 20 (66.7%)

>CR2 1 (4.3%) 4 (13.3%)In relapse 0 0

Follow up Minimum (months) 0.4 117.4Maximum (months) 106.4 272.9Median (months) 56.7 193.4

Death 4 6Loss to follow up 0 05-year overall survival 82.6% 80%5-year event free survival 75.1% 66.7%

Outcomes of Paediatric APL in HK206

B. Clinical features (see Table 3)Apart from typical presentation of pancytopenia as

in other subtypes of AML (except microgranular APLwhich tends to present with higher peripheral blastcounts) (see Table 3 case 50), APL has uniquepresen ta t ion of b leed ing secondary to severecoagulopathy (especially for hypergranular APL variant)contributed by both disseminated intravascularcoagulation (DIC) and primary fibrinolysis, which cancause life-threatening pulmonary or cerebrovascularhaemorrhage (up to 40 percent) and even earlyhaemorrhagic death (10 to 20 percent).11,12 In localpaediatric APL cohort, clinical bleeding features(conjunctival or retinal haemorrhage, intracranial orsubdural haemorrhage, oral mucosal bleeding, gumbleeding, epistaxis, bruising, petechiae, pulmonaryhaemorrhage, menorrhagia, haematuria, per rectalbleeding or haemarthrosis) was noted in 85 percent (39out of 46) of patients on initial presentation. In ourlocality, 4 out of 53 patients (7.5 percent) died ofhaemorrhagic complications before initiation oftreatment, all due to massive intracerebral haemorrhage(ICH). Infective symptoms (fever, coryzal symptomsor gastroenteritis), anaemic symptoms (pallor, fatigue,malaise, dizziness, palpitation, shortness of breath) andconstitutional symptoms (loss of appetite, weight loss)and were noted in 37 percent (17 out of 46), 24 percent(11 out of 46) and 17 percent (8 out of 46) respectively.Other clinical presentations include gum hypertrophy,bone pain, hepatomegaly and splenomegaly. It is ofinterest to note that one of the patients present withfever and diarrhoea which later evolve into intestinalobstruction, appendicitis with gangrene formation,sepsis and disseminated intravascular coagulation (DIC)(see Table 3 Case 51) . Acute colonic pseudo-obstruction (ACPO)(Ogilvie's syndrome) had beenreported in several case reports or small series13-16 tobe associated with acute myeloid (including APL) andlymphoblastic leukaemia during induction chemotherapypresumably due to use of vincristine which can causeautonomic neuropathy, but its association with APL oninitial presentation before starting chemotherapy has notbeen described in literature. As general paediatriciansor practitioners, it is important to recognise the clinicalpresentations of this rare, potentially lethal yet highlytreatable disease entity as prompt referral to paediatriconcology specialists for diagnosis and treatment couldsignificantly improve patient outcome as repeatedlyreiterated in this article.

C. Diagnosis and Pathologic Features(a) Morphology in marrow and peripheral blood smear

APL is characterised by the presence of atypicalpromyelocytes in bone marrow and/or peripheralblood. Promyelocytes are large (usually larger than20 microns in diameter) myeloid precursors ofvariable morphologies, usually with high nucleus tocytoplasmic ratio, fine chromatin, and prominentnucleoli. Cardinal feature of promyelocyte is thepresence of numerous violet cytoplasmic granules ina dense or coarse pattern obscuring the nucleus.

(b) Conventional karyotypingConventional karyotyping is an essential part of the

standard workup. It is highly specific but time-consuming (takes approximately two days for results).Around two-thirds (32/53) of local paediatric APLpatients have cytogenetics result available. APL ischaracterised by translocation t(15;17)(q22;q21)involving fusion of gene encoding the retinoic acidreceptor alpha (RARα) and promyelocytic leukaemia(PML).17,18 In local paediatric APL cohort, t(15;17)is demonstrated in 80 percent (26 out of 32 patients)with cytogenetics results available (see Table 3).PML-RAR fusion gene predicts a favourable responseto molecularly targeted therapies ATRA and arsenictrioxide (ATO).

(c) Molecular geneticsDetermination of the underlying molecular fusion

is critical for the appropriate clinical management ofAPL. Approximately 10% of APL cases with anunderlying PML-RAR fusion have no detectablet(15;17) at diagnosis, either due to unsuccessfulcytogenetic analysis, cryptic rearrangements or morecomplex rearrangements including simple varianttranslocations.19,20 Patients with molecularly but notcytogenetically detected PML-RAR fusion have anequally favourable outcome to those with overtt(15;17).21 Four out of thirty-two local paediatric APLpatients showed normal cytogenetics. Cryptictranslocation was detected by reverse transcriptasepolymerase chain reaction (RT-PCR) or fluorescencein situ hybridisation (FISH) (see Table 3).

It is important to define PML-RAR isoform as abasis for subsequent MRD monitoring. In minority ofcases (<5%), RARα is fused to alternative partners,commonest being nucleophosmin (NPM1) resultingfrom the t(5;17)(q35;q21) translocation20,22 and NuMAas a result of t(11;17)(q13;q21). These rare paediatricretinoid-sensitive subtypes of APL, with NPM1-

Chan et al 207

RARA and NuMA-RARA fusions,20,22,23 may not besensitive to ATO3; whereas APL involving PLZF andSTAT5b as a result of t(11;17)(q23;q21) and interstitialdeletion of chromosome 17 respectively, are resistantto ATRA3. These fusions were not identified in localcohort.

(d) FMS-l ike tyros ine k inase- in ternal tandemduplications (FLT3-ITD)

Among local cohort, FLT3-ITD was detected in onepatient, but its prognostic significance is uncertain.24,25

(see Table 3)(e) Immunophenotyping

APL phenotype is considered CD34 negative/partialor weak positive, HLA-DR negative, CD13 positive,CD33 positive, CD11b negative, CD15 weak ornegative, CD117 weak/variable, and sometimes CD2positive and CD56 positive.

D. Treatment(a) Drawbacks of anthracycline-based protocol

(HKPHOSG AML 1996)In Hong Kong, HKPHOSG had adopted the

HKPHOSG AML 96 protocol based on UK MRCAML 10/12 protocol in treating territory-widepaediatric APL patients in all local paediatriconcology centres during 1997 to 2008. While theoutcome is generally favourable with 5-year OS of80%, 5-year EFS is just 66.7% with relapse rate of16.7%. This drawback is also being identified fromoverseas experience.26-28 Using anthracycline−basedchemotherapy protocols with ATRA followed bymaintenance with ATRA, 6-mercaptopurine (6-MP)and methotrexate (MTX), relapse rates for patientswith a WBC <10x109/L at diagnosis are typically 10%or less, while rates may exceed 20% for patients witha WBC ≥10x109/L, which echoed with local situation.

(b) Rationale of changes in ICC APL 2001 protocolICC APL 2001 protocol based on the results of

AIEOP-AIDA 93 trial demonstrating promising resultswith combined use of ATRA and anthracyclines.GIMEMA-AIEOP AIDA 93 study recruited 983patients with newly diagnosed APL between January1993 and June 2000 of whom 124 (13%) were agedless than 18 years. Induction chemotherapy consistedof Idarubicin 12 mg/m2/day x 4 given with ATRA 25mg/m2/day (compared to an adult dose of 45 mg/m2/day) until remission or maximum of 90 days.Consolidation consisted of 3 monthly courses:Cytarabine (Ara-C) 1 gm/m2/day x 4 with Idarubicin5 mg/m2/day x 4 (course 1), Mitoxantrone 10 mg/m2/

day x 5 with etoposide (VP-16) 100 mg/m2/day x 5(course 2), and Idarubicin 12 mg/m2/day x 1, Ara-C150 mg/m2/8 hourly x 5 and 6-thioguanine (6-TG)70 mg/m2/8 hourly x 5 (course 3). Molecularresponse by RT-PCR was assessed following the thirdconsolidation course and PCR-negative patients wereinitially randomised to one of four maintenance arms(6-MP+MTX; ATRA alone; ATRA+6-MP+MTX; nomaintenance) but subsequently to ATRA alone orATRA+6-MP+MTX. One hundred and three (96%) ofthe 107 evaluable children achieved haematologicalcomplete remission (CR) and 4 died during induction(1 sepsis; 3 intracranial haemorrhage). Overt ATRAsyndrome occurred in two cases and pseudotumorcerebri in 10 children. The OS, disease-free-survival(DFS) and EFS, at more than 10 years, were 89%,78% and 76% respectively. The only significantpredictor of haematological CR was the WBC; 100%vs 89% (p=0.01) for WBC less than 10x109/L andequal to or greater than 10x109/L respectively. A WBCat diagnosis of equal to or greater than 10x109/L alsohad a significant impact on EFS (59% vs 83% at 10years). Ninety-four children were evaluable for RT-PCR at the end of consolidation; of these, 91 (97%)tested PCR-negative and 3 PCR-positive. Childrenwho were in molecular remission after consolidationtherapy and randomised to receive ATRA andmaintenance chemotherapy were significantly morelikely to remain in molecular remission than thosechildren who received ATRA alone (77% vs 42%respectively; p=0.0177).

GIMEMA-AIEOP AIDA 93 study reduced dose ofATRA as compared to adults in view of the apparentincrease in ATRA-related side effects observed inchildren. Besides, cumulative dose of cardiotoxicanthracyclines is also lowered. Cytarabine isdelivered in consolidation assuming its role inpatients with high WBC, which is a commonersituation in childhood APL, though its benefitremained unproven.29,30 The combination ofintermittent ATRA and chemotherapy may reduce ratesof molecular and frank relapse and hence improveoutcome, particularly for patients with a high whitWBC at presentation.30-33 With ICC APL 2001protocol, superior outcomes had been demonstratedwith 5-year OS and EFS of 82.6% and 75.1%respectively with a median follow up period of 56.7months for local cohort.

(c) Minimal residual disease (MRD) directed pre-emptive therapy

Outcomes of Paediatric APL in HK208

(continued on page 209)

Table 3 Clinical features, cytogenetic results, treatment protocols used and outcomes of local paediatric APL cases (1996-2017)

Case Sex Age Initial clinical presentation Cytogenetics Treatment protocol Outcome

1 F 9.5 Not available Not available HKPHOSG AML 1996 Dead, CR1 (ALLObmt)

2 F 12.83 Not available Not available HKPHOSG AML 1996 Alive, CR2 (AUTObmt)

3 M 14.92 Not available Not available Not applicable Dead (ICH)

4 M 13 Not available Not available HKPHOSG AML 1996 Alive, CR1

5 F 3.42 Not available Not available HKPHOSG AML 1996 Alive, CR1(ALLObmt)

6 F 10.33 Not available Not available HKPHOSG AML 1996 Dead (ICH)

7 F 7.75 Not available Not available HKPHOSG AML 1996 Alive, CR1

8 F 14.25 Not available Not available HKPHOSG AML 1996 Alive, CR1

9 F 14.08 Bilateral knee and ankle pain, Not available HKPHOSG AML 1996 Alive, CR3bruising, left ICH and right hemiplegia

10 F 13 Epistaxis, bruises, lethargy, pallor Not available HKPHOSG AML 1996 Alive, CR1

11 F 9.5 Petechiae, gum bleeding, dizziness, Not available HKPHOSG AML 1996 Alive, CR3palpitation, hepatosplenomegaly (AUTObmt)

12 M 15.5 PRB from haemorrhoid Not available HKPHOSG AML 1996 Alive, CR1(promyelocytes noted in peripheral blood)

13 M 7.08 Gum bleeding and hypertrophy, Not available HKPHOSG AML 1996 Death in inductioneasy bruising, malaise (right parietal ICH, DIC)

14 M 12 Fever, easy bruising, Not available HKPHOSG AML 1996 Relapsed, death ingum bleeding, right wrist (developed ATRA syndrome) induction (massive left

and knee pain parieto-occipital ICH)

15 M 7.67 Recurrent epistaxis, gum bleeding, Not available HKPHOSG AML 1996 Alive, CR1pallor, petechiae, coryzal symptoms

16 M 8.17 Gum bleeding, bruising Not available HKPHOSG AML 1996 Alive, CR1

17 M 12.08 Gum bleeding, fever, poor appetite Not available HKPHOSG AML 1996 Alive, CR1

18 M 13.58 Gum bleeding, petechiae, visual Not available HKPHOSG AML 1996 Alive, CR1blurring, fever without obvious focus (developed ATRA syndrome)

19 F 9.83 Menorrhagia, easy bruising 46,XX[16] HKPHOSG AML 1996 Alive, CR1cryptic t(15;17) detected by FISH

20 M 4.33 Persistent fever, coryzal 46,XY,add(15)(q22)[6]/ HKPHOSG AML 1996 Alive, CR1symptoms, gum bleeding, 46,idem,add(2)(p13)[7]/

bruises, weight loss 46,XY[2]

21 M 4.58 Fever, coryzal symptoms, 46,XY,t(15;17) HKPHOSG AML 1996 Alive, CR1poor appetite, easy bruising (q22;q12)[6]

22 M 15.17 Prolonged fever, coryzal Not available Not applicable Died before treatmentsymptoms, headache, dizziness,

repeated vomiting, bruises

23 F 15.92 Prolonged fever, pallor, weight loss 46,XX,t(15;17) HKPHOSG AML 1996 Alive, CR1No clinical bleeding (q22;q12)[6]/46,XX[1]

24 F 8.33 Gum swelling, bruises 46,XX,t(15;17)(q22;q12)[18] HKPHOSG AML 1996 Alive, CR1

25 F 7.75 Bruises, weight loss Not available HKPHOSG AML 1996 Alive, CR1

Chan et al 209

(continued on page 210)

Table 3 Clinical features, cytogenetic results, treatment protocols used and outcomes of local paediatric APL cases (1996-2017) (cont'd)

Case Sex Age Initial clinical presentation Cytogenetics Treatment protocol Outcome

26 M 7.08 Not available Not available HKPHOSG AML 1996 Alive, CR1

27 M 13.83 Prolonged fever, gum 46,XY,t(15;17) HKPHOSG AML 1996 Alive, CR1bleeding, bruises, PRB, pallor, dizziness (q22;q12)[10]

28 F 2.83 Fever, reduced appetite, nausea 46,XX,t(15;17) HKPHOSG AML 1996 Alive, CR1No clinical bleeding (q22;q21)[20]

29 M 13 Recurrent bruises, 46,XY,t(15;17) HKPHOSG AML 1996 Alive, CR1gum bleeding (q22;q21)[16]/46,XY[4]

30 F 13.17 Menorrhagia, recurrent gum 46,XX,t(15;17) HKPHOSG AML 1996 Alive, CR1bleeding, bruising, hepatomegaly, (q22;q21)[20]

retinal haemorrhage

31 F 11.42 Pallor, fever, cough, bone pain, 46,XX[20] HKPHOSG AML 1996 Alive, CR1bilateral retinal haemorrhage cryptic t(15;17)

detected by FISH

32 M 12.17 Low grade fever, coryzal 47,XY,t(2;19) HKPHOSG AML 1996 Alive, CR1symptoms, recurrent epistaxis, (p12;p13),t(15;17)

right retinal haemorrhage (q22;q21),+21[6]/46,XY[4]

33 M 3.67 Prolonged fever, coryzal 46,XY,t(15;17) ICC APL 2001 Alive, CR1symptoms, pallor, RSV pneumonia (q22;q12)[7]/

No clinical bleeding 46,XY[1]

34 F 12.17 Menorrhagia, bruising, 46,XX,t(15;17) ICC APL 2001 Death in inductiongum bleeding, malaise, headache (q22;q12)[6] (bilateral severe ICH)

35 F 17.75 Bruising, petechiae 46,XX,t(15;17)(q22;q12)[10] ICC APL 2001 Alive, CR1

36 M 9 Petechiae, gum bleeding/ Not available ICC APL 2001 Alive, CR1hypertrophy (diagnosed in Nepal)

37 F 16.08 Menorrhagia, bilateral lower 46,XX,t(15;17) ICC APL 2001 Alive, CR1limb bruising, gum bleeding, (q22;q12)[13]/ (developed ATRA syndrome)

malaise, SOB 46,XX[2]

38 F 16.08 Menorrhagia, bruising, malaise, 46,XX,t(15;17) ICC APL 2001 Alive, CR1palpitation, blurring of vision (q22;q12)[6]

39 M 5.42 Prolonged fever, coryzal and 46,XY,t(15;17) ICC APL 2001 Alive, CR1GE symptoms, retinal haemorrhage, (q22;q12)[14]/

gum hypertrophy, leg pain, mucosal bleed 46,XY[4]

40 M 16.5 Pallor, malaise, weight loss, 46,XY,t(15;17) ICC APL 2001 Alive, CR1bruising, petechiae (q22;q12)[18]/46,XY[2]

41 M 14.5 Central chest pain, gum 46,XY,add(6)(q11), ICC APL 2001 Death in inductionbleeding, petechiae t(15;17)(q22;q12)[16]/ (multifocal ICH/SDH,

46,XY[3] refractory seizures,

pulmonary haemorrhage)

42 F 14.83 Fever, epigastric discomfort, 46,XX,t(15;17) ICC APL 2001 Alive, CR1vomit, left thigh bruising, (q22;q12)[11]

bilateral conjunctival haemorrhage

Outcomes of Paediatric APL in HK210

Table 3 Clinical features, cytogenetic results, treatment protocols used and outcomes of local paediatric APL cases (1996-2017)

Case Sex Age Initial clinical presentation Cytogenetics Treatment protocol Outcome

43 M 7.58 Pallor, left eye and bilateral 46,XY,t(15;17) ICC APL 2001 Alive, CR1lower limb bruising (q22;q12)[13]/47,idem,+8[7]

44 M 15.5 Fever, chills and rigour, No growth ICC APL 2001 Alive, CR1headache, vomiting (developed ATRA syndrome)

No clinical bleeding

45 M 13.67 Prolonged fever, coryzal 46,XY,t(15;17) ICC APL 2001 Alive, CR1symptoms, bruises (q22;q12)[19]

46 F 3.5 Pallor, bruises, petechiae 46,XX,i(17)(q10)[3]/46,XX[20] ICC APL 2001 Alive, CR1

47 M 11.33 Prolonged fever, coryzal 46,XY,der(14)t(9;14) ICC APL 2001 Alive, CR1symptoms, vomiting (q13;q32),t(15;17)

No clinical bleeding (q22;q12)[20]

48 F 13.5 Poor appetite, bruising, 46,XX,t(15;17) ICC APL 2001 Alive, CR1gum bleeding, petechiae, pallor (q22;q12)[11]

49 F 15.33 Right SDH with 46,XX,t(15;17)(q22;q12) ICC APL 2001 Died before treatmentincreased ICP and coning [5]/46,idem,+21,rob(21;21) (brainstem death, DI)

(q10;q10)[9]/46,XX[5]

50 F 2.42 Fever without localizing symptoms 46,XX,t(15;17) ICC APL 2001 Alive, CR1or signs, hyperleukocytosis (q24;q21)[20]

No clinical bleeding

51 F 12.83 Fever and diarrhoea evolved into 46,XX,t(15;17) ICC APL 2001 Alive, CR1intestinal obstruction, appendicitis and (q22;q12)[10]

gangrene formation, sepsis, DICNo clinical bleeding

52 M 12.75 Intermittent fever, 46,XY[15] ICC APL 2001 Alive, CR1gum bleeding, bruising cryptic t(15;17) detected

by FISH, FLT3-ITD positive

53 M 17.25 Fever, petechiae, bilateral 46,XY,t(15;17) ICC APL 2001 Alive, CR1conjunctival and retinal haemorrhage, (q24;q21)[19]/ (developed ATRA syndrome)

left calf swelling and pain, 46,XY[2]headache, left temporal ICH

54 M 7.5 Prolonged fever, coryzal symptoms, 46,XY,t(15;17) ICC APL 2001 Alive, CR1gum bleeding, petechiae, (q24;q21)[3]/46,idem,

pancytopenia del(7)(q32q36)[17]/46,XY[1]

55 M 11.25 Multifocal ICH and 46,XY[20] ICC APL 2001 Died before treatmentSAH, tonsillar herniation cryptic t(15;17) (AKI, DIC, sepsis,

detected by FISH gangrene of extremities)

Note: AKI: acute kidney injury; ALLObmt: allogenic transplant; AUTObmt: autologous transplant; DI: diabetes insipidus; DIC: disseminated intravascularcoagulation; FISH: fluorescence in situ hybridisation; GE: gastroenteritis; ICH: intracranial haemorrhage; ICP: intracranial pressure; PRB: per rectal bleeding;SAH: subarachnoid haemorrhage; SDH: subdural haemorrhage; SOB: shortness of breath

Chan et al 211

Figure 1 Roadmap for treatment of paediatric APL using HKPHOSG AML 1996 protocol (based on UK MRC AML protocol 10/12).

Note: All trans retinoic acid (ATRA) will be given as single oral agent at 45 mg/m2/day in daily single dose. The maximum duration oftreatment is 6 weeks. Once complete remission is achieved, the AML chemotherapy protocol should be started. Combination chemotherapywith induction chemotherapy (MAE) together with ATRA should be given if: (a) WBC at diagnosis >10x109/L; (b) WBC rise to >20x109/Lduring ATRA treatment; (c) presence of retinoic acid syndrome. ATRA should be continued until clinical remission or maximum of 90 days.

Monitoring of minimal residual disease (MRD) hasbeen shown to be an independent risk factor forrelapse34 as demonstrated in studies from Italy28,35-37

and the United Kingdom.21 German AML CooperativeGroup also demonstrated that patients who failed toachieve a 3-log reduction in PML-RARα transcriptlevel within the first 3-4 months of therapy asdetected by quantitative RT-PCR had an increased riskof early relapse.38

Pre-emptive therapy for patients with positive MRDimpending relapse confers survival benefit thanretreating patient when they have progressed to frankrelapse due to the well-recognised risk of inductiondeath due to associated coagulopathy as evidenced bythe GIMEMA and PETHEMA groups.36,39,40 Patientswith persistent PCR positivity or molecular relapsewill receive molecularly targeted therapy, with ATOand may proceed to stem cell transplantation. Forpatients achieving molecular remission and with a PCRnegative stem cell harvest, good results have been

achieved with autologous stem cell transplantation.41,42

Allogeneic transplant is generally reserved for APLpatients in whom harvest of PCR negative stem cellsis unsuccessful or who fail to achieve secondmolecular remission.

(d) Arsenic Trioxide (ATO)Therapeu t i c u se o f ATO has we l l been

documented in treatment of various diseases acrosscenturies despite its historical reputation as a toxinand a poison. ATO has now emerged as a treatmentof choice for patients with refractory/relapsedPML-RAR+ APL.43,44

Although second remission could be achievedthrough re-induction with ATRA alone or incombination with chemotherapy, 2 cycles of ATO atrecommended dose of 0.15 mg/kg/day enablesinduction of molecular remission in majority ofpatient without significant myelosuppression,43 whichallows outpatient-based therapy and reduction inanthracycline exposure.

Outcomes of Paediatric APL in HK212

Figure 2 (a) Roadmap for treatment of paediatric APL using ICC APL Study 2001 protocol. (b) Roadmap for treatment of paediatric APL(including relapsed and primary refractory APL) using ICC APL Study 2001 protocol.

(a)

(b)

Chan et al 213

Figure 3 Overall and event-free survival of paediatric APL in HK (HKPHOSG AML 1996 based on UK AML protocol 10/12 vs ICCAPL 2001 protocol).

The optimal dosing schedule for ATO is not known.At least three dosing regimens are currently in use:(i) continuous dosing at 0.15 mg/kg/day, (ii)intermittent dosing at 0.15 mg/kg/day for 5 days, offfor two days, weekly x 4 with a two week break whendaily ATRA is given and the cycle repeated x 2 and(iii) intermittent dosing starting with a loading doseof 0.3 mg/kg for 5 days followed by 0.25 mg/kg twiceweekly. These dosing schedules are comparable interms of efficacy and toxicity profiles.45,46 The ICC-APL Study 2001 protocol is based on the conventionalcontinuous dosing schedule recommended by themanufacturer, Cephalon. However, alternativeequivalent dosing schedules may be preferred forlogistic reason. ATO has also been evaluated as partof first-line therapy of APL in the recent USIntergroup trial C971042. Safety and effectivenessin paediatric patients under 5 years of age has not beenstudied to any great extent. From personalcommunications with local adult haematologists,promising result had been demonstrated with the useof oral arsensol in adult APL patients in Hong Kong.Usage of oral arsenol has been proposed as amonotherapy in selected low risk paediatric APLpatients, sparing the use of anthracyclines. Futurestudies could be proposed related to this.

(e) ATRA syndrome / retinoic acid syndrome (RAS) /differentiation syndrome

Despite high efficacy of ATRA and ATO in treatingAPL patients, their usage could lead to retinoic acid

syndrome (RAS) manifested as fever, dyspnoea,weight gain, hypotension and pulmonary infiltration.The mechanism of RAS is unknown but it is speculatedthat capillary leak syndrome due to cytokine releasefrom differentiating myeloid cells and mediation bycathepsin G could play a role.47 In local cohort, nearlyone-tenth (5 out of 53 cases) developed RAS afterATRA/ATO. Symptoms usually subside after ATRA/ATO is d i scont inued in conjunc t ion wi thadministrat ion of short course intravenousdexamethasone as in all 5 local cases.

(f) Gemtuzumab Ozogamicin (Mylotarg®)Gemtuzumab ozogamicin (GO) is an anti-CD33

targeted therapy with significant anti-leukaemicactivity against APL due to high CD33 expression inAPL cells.48-50 Its conjugate contains the intercalatinganthracycline calicheamicin, which links the antibodyto the cytotoxic agent, enables calicheamicin to bereleased only intracellularly, thereby avoiding muchof the anthracycline-related toxicity. Its major side-effects include myelosuppression, allergic reactionsand sinusoidal obstruction syndrome (SOS).

Strengths and Limitations

This territory-wide retrospective study summarisedinitial clinical presentations, treatment responses, clinicaloutcomes and complications HK children and adolescentsat age 18 or below suffering from this rare subtype of APL

Outcomes of Paediatric APL in HK214

in the last 2 decades. It also made comparisons betweenthe two previously adopted protocols (HKPHOSG AML1996 versus ICC APL 2001), which consolidated localclinical experience from 5 children oncology centres inHK. The intrinsic limitation of this study is that the twogroups compared are at different time points (1997-2008vs 2009 till present). The follow up time for the newerprotocol is much shorter. The secular trend ofimprovement is hard to be discerned whether ATRA-basedtherapy demonstrates better outcome than AML-basedchemotherapy in treating paediatric APL patients is due totreatment itself as there could be variants and confounderslike the team's learning curve, expertise mix, supportivecare by intensivists as well as multidisciplinary nursing andpharmacy team care which could potentially led to betteroutcome.

Conclusion

Retrospective review of local paediatric APL cohortreveals that ATRA-based therapy demonstrate betteroutcome than AML-based chemotherapy in treatingpaediatric APL patients. With the establishment of HKCH,it is hoped that a unified updated treatment protocol incollaboration with adult haematologists and internationalcentres would be formulated, with incorporation of moreaccurate and wider panel of molecular diagnostics. Besides,it is anticipated that with the newly formed hub-and-spokeclinical service model, prompt referral of suspected ordiagnosed paediatric APL cases from regional hospitals toHKCH could be enhanced.

Declaration of Interest

All the authors report no conflicts of interest.

Acknowledgements

The authors would like to thank HKPHOSG data officersfor data accrual and the medical, nursing and pharmacy teaminvolved in patient care.

References

1. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to theWorld Health Organization classification of myeloid neoplasms and

acute leukemia. Blood 2016; 127:2391-405.2. Asou N, Adachi K, Tamura J, et al. Analysis of prognostic factors

in newly diagnosed acute promyelocytic leukemia treated with all-trans retinoic acid and chemotherapy. Japan Adult Leukemia StudyGroup. J Clin Oncol 1998;16:78-85.

3. Beaumont M, Sanz M, Carli P, et al. Therapy-related acutepromyelocytic leukemia. J Clin Oncol 2003;21:2123-37.

4. Yamamoto JF, Goodman MT. Patterns of leukemia incidence inthe United States by subtype and demographic characteristics,1997-2002. Cancer Causes Control 2008;19:379-90.

5. Dores GM, Devesa SS, Curtis RE, et al. Acute leukemia incidenceand patient survival among children and adults in the United States,2001-2007. Blood 2012; 119:34.

6. Matasar MJ, Ritchie EK, Consedine N, Magai C, Neugut AI.Incidence rates of acute promyelocytic leukemia among Hispanics,blacks, Asians, and non-Hispanic whites in the United States. EurJ Cancer Prev 2006;15:367-70.

7. Sanz MA, Grimwade D, Tallman MS, et al. Management of acutepromyelocytic leukemia: recommendations from an expert panelon behalf of the European LeukemiaNet. Blood 2009;113:1875-91.

8. Chen Y, Kantarjian H, Wang H, Cortes J, Ravandi F. Acutepromyelocytic leukemia: a population-based study on incidenceand survival in the United States, 1975-2008. Cancer 2012;118:5811-8.

9. Park JH, Qiao B, Panageas KS, et al. Early death rate in acutepromyelocytic leukemia remains high despite all-trans retinoic acid.Blood 2011;118:1248-54.

10. Beaumont M, Sanz M, Carli PM, et al. Therapy-related acutepromyelocytic leukemia. J Clin Oncol 2003;21:2123-37.

11. Fenaux P, Le Deley MC, Castaigne S, et al. Effect of alltransretinoic acid in newly diagnosed acute promyelocytic leukemia.Results of a multicenter randomized trial. European APL 91 Group.Blood 1993;82:3241-9.

12. Rodeghiero F, Avvisati G, Castaman G, Barbui T, Mandelli F. Earlydeaths and anti-hemorrhagic treatments in acute promyelocyticleukemia. A GIMEMA retrospective study in 268 consecutivepatients. Blood 1990;75:2112-7.

13. Wells CI, O'Grady G, Bissett IP. Acute colonic pseudo-obstruction:A systematic review of aetiology and mechanisms. World JGastroenterol 2017;23:5634-44.

14. Jessop M, Choo K, Little M. Acute colonic pseudo-obstruction inpaediatric oncology patients. J Paediatr Child Health 2010;46:698-9.

15. Kurtz J, Andres E, Maloisel F, et al. Acute colonic pseudoobstructionin acute promyelocytic leukemia. Gastroenterol Clin Biol 1997;21:629-30.

16. Park JA, Yun JH, Kang HJ, Shin HY, Ahn HS. Acute intestinalpseudo-obstruction after induction treatment of relapsed acutepromyelocytic leukemia with arsenic trioxide. Pediatr Blood Cancer2008;50:872-4.

17. Melnick A, Licht JD. Deconstructing a disease: RARalpha, itsfusion partners, and their roles in the pathogenesis of acutepromyelocytic leukemia. Blood 1999;93:3167-215.

18. Kakizuka A, Miller WH Jr, Umesono K, et al. Chromosomaltranslocation t(15;17) in human acute promyelocytic leukemia fusesRAR alpha with a novel putative transcription factor, PML. Cell1991;66:663-74.

19. Grimwade D, Gorman P, Duprez E, et al. Characterization of crypticrearrangements and variant translocations in acute promyelocyticleukemia. Blood 1997;90:4876-85.

Chan et al 215

of the PML/RARalpha fusion gene in patients with acutepromyelocytic leukemia enrolled in the GIMEMA-AIEOPmulticenter "AIDA" trial. Blood 1998;92:784-9.

36. Lo Coco F, Diverio D, Avvisati G, et al. Therapy of molecularrelapse in acute promyelocytic leukemia. Blood 1999;94:2225-9.

37. Lo Coco F, Romano A, Mengarelli A, et al. Allogeneic stem celltransplantation for advanced acute promyelocytic leukemia: resultsin patients treated in second molecular remission or withmolecularly persistent disease. Leukemia 2003;17:930-3.

38. Schnittger S, Weisser M, Schoch C, et al. New score predictingfor prognosis in PML-RARA+, AML1-ETO+, or CBFBMYH11+acute myeloid leukemia based on quantification of fusion transcripts.Blood 2003;102:2746-55.

39. Esteve J, Escoda L, Martin G, et al. Outcome of patients withacute promyelocytic leukemia failing to front-line treatment withall-trans retinoic acid and anthracycline-based chemotherapy(PETHEMA protocols LPA96 and LPA99): benefit of an earlyintervention. Leukemia 2007;21:446-52.

40. Lo Coco F, De Santis S, Esposito A, Divona M, Diverio D.Molecular monitoring of hematologic malignancies:current andfuture issues. Semin Hematol. Semin Hematol 2002;39(2 Suppl1):14-7.

41. Meloni G, Diverio D, Vignetti M, et al. Autologous bone marrowtransplantation for acute promyelocytic leukemia in secondremission:prognostic relevance of pretransplant minimal residualdisease assessment by reverse-transcription polymerase chain reactionof the PML/RAR alpha fusion gene. Blood 1997;90:1321-5.

42. Thomas X, Dombret H, Cordonnier C, et al. Treatment of relapsingacute promyelocytic leukemia by all-trans retinoic acid therapyfollowed by timed sequential chemotherapy and stem celltransplantation. Leukemia 2000;14:1006-13.

43. Soignet SL, Frankel SR, Douer D, et al. United States multicenterstudy of arsenic trioxide in relapsed acute promyelocytic leukemia.J Clin Oncol 2001;19:3852-60.

44. Niu C, Yan H, Yu T, et al. Studies on treatment of acutepromyelocytic leukemia with arsenic trioxide: remission induction,follow-up, and molecular monitoring in 11 newly diagnosed and47 relapsed acute promyelocytic leukemia patients. Blood 1999;94:3315-24.

45. Grimwade D, Jovanovic J, Diverio D, et al. Real-time detection ofPML-RARA and RARA-PML fusion transcripts in high risk acutepromyelocytic leukemia (APL) treated with arsenic trioxide (ATO):Implications for realization of pre-emptive therapy for molecularrelapse. Am Soc Hematol 2006.

46. Fox E, Razzouk BI, Widemann BC, et al. Phase 1 trial andpharmacokinetic study of arsenic trioxide in children and adolescentswith refractory or relapsed acute leukemia, including acutepromyelocytic leukemia or lymphoma. Blood 2008;111:566-73.

47. Tallman MS. Retinoic acid syndrome: a problem of the past?Leukemia 2002;16:160-1.

48. Lo Coco F, Ammatuna E, Noguera N. Treatment of acutepromyelocytic leukemia with gemtuzumab ozogamicin. Clin AdvHematol Oncol 2006;4:57-62.

49. Lo-Coco F, Cimino G, Breccia M et al. Gemtuzumab ozogamicin(Mylotarg) as a single agent for molecularly relapsed acutepromyelocytic leukemia. Blood 2004;104:1995-9.

50. Petti MC, Pinazzi MB, Diverio D, et al. Prolonged molecularremission in advanced acute promyelocytic leukaemia aftertreatment with gemtuzumab ozogamicin (MylotargTM CMA-676).Br J Haematol 2001;115:63-5.

20. Grimwade D, Biondi A, Mozziconacci MJ, et al. Characterizationof acute promyelocytic leukemia cases lacking the classic t (15;17): results of the European Working Party. Blood 2000;96:1297-308.

21. Burnett AK, Grimwade D, Solomon E, Wheatley K, GoldstoneAH. Presenting white blood cell count and kinetics of molecularremission predict prognosis in acute promyelocytic leukemiatreated with all-trans retinoic acid:result of the Randomized MRCTrial. Blood 1999;93:4131-43.

22. Redner RL, Rush EA, Faas S, Rudert WA, Corey SJ. The t(5;17)variant of acute promyelocytic leukemia expresses a nucleophosmin-retinoic acid receptor fusion. Blood 1996;87:882-6.

23. Wells RA, Catzavelos C, Kamel-Reid S. Fusion of retinoic acidreceptor alpha to NuMA, the nuclear mitotic apparatus protein,by a variant translocation in acute promyelocytic leukaemia. Naturegenetics 1997;17:109-13.

24. Poire X, Moser BK, Gallagher RE, et al. Arsenic trioxide in front-line therapy of acute promyelocytic leukemia (C9710): prognosticsignificance of FLT3 mutations and complex karyotype. LeukLymphoma 2014;55:1523-32.

25. Kutny MA, Moser BK, Laumann K, et al. FLT3 mutation status isa predictor of early death in pediatric acute promyelocytic leukemia:a report from the Children's Oncology Group. Pediatr Blood Cancer2012;59:662-7.

26. de Botton S, Coiteux V, Chevret S, et al. Outcome of childhoodacute promyelocytic leukemia with all-trans-retinoic acid andchemotherapy. J Clin Oncol 2004;22:1404-12.

27. Ortega JJ, Madero L, Martin G, et al; PETHEMA Group Treatmentwith all-trans retinoic acid and anthracycline monochemotherapyfor children with acute promyelocytic leukemia: a multicenter studyby the PETHEMA Group. J Clin Oncol 2005;23:7632-40.

28. Testi AM, Biondi A, Lo Coco F, et al. GIMEMA-AIEOPAIDAprotocol for the treatment of newly diagnosed acute promyelocyticleukemia (APL) in children. Blood 2005;106:447-53.

29. Sanz MA, Lo Coco F, Martin G, et al. Definition of relapse riskand role of nonanthracycline drugs for consolidation in patientswith acute promyelocytic leukemia: a joint study of the PETHEMAand GIMEMA cooperative groups. Blood 2000;96:1247-53.

30. Sanz MA, Martin G, Rayon C, et al. A modified AIDA protocolwith anthracycline-based consolidation results in high antileukemicefficacy and reduced toxicity in newly diagnosed PML/ RARalpha-positive acute promyelocytic leukemia. PETHEMA group. Blood1999;9:3015-21.

31. Tallman MS, Nabhan C, Feusner JH, Rowe JM. Acutepromyelocytic leukemia: evolving therapeutic strategies. Blood2002;99:759-67.

32. Fenaux P, Chastang C, Chevret S, et al. A randomized comparisonof all transretinoic acid (ATRA) followed by chemotherapy andATRA plus chemotherapy and the role of maintenance therapy innewly diagnosed acute promyelocytic leukemia. The European APLGroup. Blood 1999;94:1192-200.

33. Sanz MA, Martin G, Lo Coco F. Choice of chemotherapy induction,consolidation and maintenance in acute promyelocytic leukaemia.Best Pract Res Clin Haematol 2003;16:433-51.

34. Grimwade D, Lo Coco F. Acute promyelocytic leukemia: a modelfor the role of molecular diagnosis and residual disease monitoringin directing treatment approach in acute myeloid leukemia.Leukemia 2002;16:1959-73.

35. Diverio D, Rossi V, Avvisati G, et al. Early detection of relapse byprospective reverse transcriptase-polymerase chain reaction analysis