»Clinical significance of mutations in the calreticulin gene (CALR) in patients with myeloproliferative neoplasms«

Uvod: V septembru 2013 je bila pri večini bolnikov, ki nimajo mutacije v genu JAK2 V61F in MPL (JAK2 V617F- in MPL-negativni bolniki) z esencialno trombocitemijo (ET) ugotovljena somatska mutacija v genu za kalretikulin (CALR). Prepoznana je bila tudi pri nekaterih bolnikih z drugimi podvrstami mieloproliferativnih novotvorb (MPN). Do sedaj je bilo prepoznanih več kot 50 različnih mutacij CALR, skoraj izključno gre za mutacije, ki povzročajo premik bralnega okvirja, in vodijo v nastanek novega C-terminalnega peptida. Glede na sodobne smernice Svetovne zdravstvene organizacije ima mutacija CALR pri ugotovitvi bolezni ET pomembno diagnostično vlogo. Diferencialna diagnoza ET je široka in poleg nekaterih drugih hematoloških bolezni vključuje različne bolezni in stanja, kot so pomanjkanje železa, okužba, vnetje, stanja po kirurških posegih ali splenektomiji. Takšna trombocitoza je sekundarna in za razliko od ET ni povezana s povečanim tveganjem za trombotično-hemoragične zaplete. Ločevanje med ET in sekundarno trombocitozo je pogosto težavno, saj ne poznamo patognomoničnega pokazatelja ET, z znanimi molekularno-genetskimi kazalci bolezni pa bolezen ugotovimo le pri približno 90 % bolnikov. V Sloveniji je določanje mutacije CALR pri bolniki s sumom na MPN v rutinski rabi šele zadnjih nekaj let, zato prisotnost mutacije CALR pri vseh naših bolnikih, ki smo jih v preteklosti obravnavali zaradi suma na MPN, ni poznana. Namen: Retrospektivno potrditi diagnostično vrednost določanja mutacije CALR pri bolnikih s sumom na MPN. Oblikovati temelje za izdelavo diagnostičnega algoritma obravnave bolnikov s trombocitozo in sumom na ET. Materiali in metode: Vključili smo JAK2 V617F-negativne bolnike, ki so bili zaradi suma na MPN napoteni v Hematološko ambulanto Univerzitetnega kliničnega centra Ljubljana med letoma 2011 in 2016. Pri vseh bolnikih smo določali prisotnost mutacij v genu CALR iz periferne krvi ali kostnega mozga. Pri tem smo uporabili presejalno metodo visokoločljivostne talilne analize, rezultate smo potrdili z elektroforezo na gelu. Tip mutacije smo natančno opredelili, pri čemer smo pri bolnikih brez najpogostejših mutacij uporabili metodo sekvenciranja po Sangerju. Po potrditvi mutacij v genu CALR smo pri bolnikih glede na razpoložljive klinične in laboratorijske podatke ponovno ugotavljali in določali bolezen. Po izvedbi preverjanja prisotnosti mutacij CALR smo med vsemi vključenimi bolniki opredelili bolnike, ki so bili v Hematološko ambulanto napoteni zaradi trombocitoze, in ob tem suma na ET. Glede na razpoložljive podatke z obravnav v Hematološki ambulanti smo bolnike razvrstili na bolnike s primarno in bolnike s sekundarno trombocitozo. Bolnike, ki imajo mutacijo CALR (CALR-pozitivne bolnike), in bolnike, ki nimajo mutacije CALR (CALR-negativne bolnike) z domnevno ET smo primerjali po pogostnosti trombotičnih zapletov. Pri bolnikih, napotenih v Hematološko ambulanto zaradi trombocitoze, smo opredelili bolnike, pri katerih je ekspertni sistem »Smart Blood Analytics« (SBA) ugotavljal bolezen s šifro D47 po Mednarodni klasifikaciji bolezni (MKB) (druge neoplazme negotovega ali neznanega značaja limfatičnega, krvotvornega in sorodnega tkiva), kamor sodi tudi ET, kot najverjetnejšo diagnozo. Ugotavljali smo občutljivost, specifičnost, pozitivno in negativno napovedno vrednost SBA pri napovedovanju ET pri bolnikih, napotenih v Hematološko ambulanto zaradi trombocitoze. Rezultati: Med 524 JAK2 V617F- negativnimi bolniki, ki so bili napoteni v Hematološko ambulanto zaradi suma na MPN, smo prepoznali 23 CALR-pozitivnih bolnikov (23/524 = 4,4 %). Pri devetih bolnikih smo retrospektivno potrdili ET. Pri dveh bolnikih smo odkrili prisotnost mutacije CALR brez premika bralnega okvirjapri nobenem od obeh bolnikov nismo diagnosticirali ET. Štiri mutacije, ki smo jih potrdili v naši raziskavi, doslej v literaturi še niso bile opisane, in so novo-ugotovljene. 237 bolnikov je bilo napotenih v Hematološko ambulanto zaradi trombocitoze, zaradi manjkajočih podatkov smo jih razvrstili 231. Primarno trombocitozo smo ugotovili pri 51 bolnikih. 42 bolnikov smo opredelili kot bolnike z domnevno ET. Pri 9 bolnikih so bile v ozadju druge hematološke bolezni, pri katerih je lahko prisotna trombocitoza. Pri 180 bolnikih je bila ugotovljena sekundarna trombocitoza. Najpogostejši vzrok sekundarne trombocitoze je bilo pomanjkanje železa. CALR-pozitivni bolniki z domnevno ET niso imeli značilno manj pogostih trombotičnih zapletov kot CALR-negativni bolniki z domnevno ET (p=0,374). Med slednjimi so bili, z izjemo enega bolnika, bolniki brez mutacij JAK2 V617F, CALR in MPL (trojno negativni bolniki). Pozitivna napovedna vrednost ekspertnega sistema SBA pri ugotavljanju ET pri bolnikih s trombocitozo je bila 23,7 %, negativna napovedna vrednost je bila 88,1 %. Občutljivost testa je bila 69 %, specifičnost je bila 50,7 %. Pri tem smo kot bolnike z ET upoštevali vse bolnike z domnevno ET. Ob upoštevanju le bolnikov s potrjeno ET po kriterijih SZO 2008 ali modificiranih kriterijih SZO je bila pozitivna napovedna vrednost ekspertnega sistema pri ugotovitvi bolezni D47 po MKB 14,8 %, negativna napovedna vrednost je bila 96,3 %. Občutljivost testa je bila 81,8 %, specifičnost je bila 50,2 %. Zaključek: Molekularno-genetska diagnostika je smiselna pri bolnikih, pri katerih je glede na anamnezo, klinični pregled in laboratorijske preiskave sum na MPN visok. V naši raziskavi smo odkrili štiri doslej še neprepoznane mutacije CALR. Večina napotitev v Hematološko ambulanto zaradi suma na ET je neustreznih. Pred napotitvijo v Hematološko ambulanto je potrebno izključiti najpogostejša stanja, ki lahko povzročajo trombocitozo. Glede na našo raziskavo je v naši populaciji najpogosteje spregledano pomanjkanje železa, ki je lahko prisotno kljub odsotnosti anemije. Trojno negativne bolnike z vztrajajočim sumom na ET bi morali zaradi primerljivega tveganja za trombotične zaplete v primerjavi s CALR-pozitivnimi bolniki obravnavati, kot da imajo ET. Te bolnike bi morali obravnavati individualno in glede na tveganje za trombotične zaplete zaščititi vsaj z acetilsalicilno kislino. SBA je manj učinkovit pri ugotavljanju ET kot hematolog.Introduction: In September 2013 somatic mutations in the calreticulin gene (CALR) were recognised in most JAK2 V617F- and MPL-negative patients with ET. CALR mutations have also been recognised in patients with other myeloproliferative neoplasm (MPN) subtypes. Up to this date, more than 50 CALR mutations have been confirmed, almost all are frameshift mutations that result in a novel C-terminal peptide. Based on the renewed World Health Organisation (WHO) criteria, CALR mutations are an important diagnostic hallmark of MPN. The differential diagnosis of ET is broad and apart from other haematological diseases includes different diseases and states such as iron deficiency, infection, inflammation, postsurgical period, hyposplenism. Thrombocytosis that results from other diseases and states is called secondary thrombocytosis and does not represent an increased risk for thrombo-haemorrhagic complications. The distinction between primary and secondary thrombocytosis is often difficult as there are no specific markers of ET. Today we can recognise the disease based on molecular-genetic markers in about 90% of all patients. In Slovenia, CALR mutations have only routinely been searched for in patients with suspected MPN in the past few years. Therefore, CALR genetic status in all patients with suspected MPN in our population in the past is unknown. Aim: To retrospectively confirm the diagnostic importance of CALR mutation confirmation in patients with suspected MPN. To form a scientific basis for the diagnostic approach to patients with suspected ET. Materials and methods: JAK2 V617F-negative patients with suspected MPN that were referred to the Department of Haematology between 2011 and 2016 were included in the study. CALR mutations were tested in all patients from blood or bone marrow samples. The screening method was »high resolution melting« analysis. All results were confirmed by gel electrophoresis. Types of CALR mutations were defined. Sanger sequencing was used in all patients’ samples with non-typical mutations. Based on CALR mutation confirmation all patients were re-diagnosed based on the available clinical and laboratory data. After CALR mutation confirmation we further analysed only patients with thrombocytosis and suspected ET. Based on the available clinical and laboratory data patients with primary and secondary thrombocytosis were defined. The number of thrombotic complications were compared in CALR-positive and CALR-negative patients with suspected ET. Artificial intelligence programme »Smart Blood Analytics« (SBA) was used to define patients with thrombocytosis whose most probable diagnosis after the initial presentation at the Department of Haematology was D47 based on International Classification of Diseases (ICD) (Other neoplasms of uncertain or unknown behaviour of lymphoid, haematopoietic and related tissue). D47 includes the diagnosis of ET. Sensitivity, specificity, positive and negative predictive values of SBA programme in diagnosing D47 in patients with thrombocytosis were defined. Results: 23 (23/524 = 4.4%) CALR-positive patients with suspected MPN were discovered. In nine patients ET was retrospectively confirmed according to modified WHO criteria. In two patients in-frame CALR mutations were confirmed, none of these patients were diagnosed with ET. We discovered four novel mutations in our study that have previously not been described. 237 patients were referred to the Department of Haematology due to thrombocytosis, due to missing data 231 patients were analysed. Primary haematological disease was confirmed in 51 patients. 42 patients were diagnosed with suspected ET and 9 patients were diagnosed with other haematological diseases. In 180 patients secondary thrombocytosis was confirmed. The most common cause of secondary thrombocytosis was iron deficiency. Thrombotic complications were not significantly less common in CALR-positive patients compared to CALR-negative patients with suspected ET (p=0.374). CALR-negative patients with suspected ET were triple negative patients (JAK2 V617F, CALR and MPL- negative) with the exception of one patient. Positive predictive value of SBA in D47 prediction was 23.7%, negative predictive value was 88.1%. The sensitivity of the test was 69% and the specificity was 50.7%. As patients with ET we included all patients with suspected ET. By including only patients with WHO 2008 or modified WHO criteria confirmed disease the positive predictive value was 14.8% and negative predictive value 96.3%. The sensitivity of the test was 81.8% and the specificity was 50.2%. Conclusion: Molecular-genetic testing should be reserved for patients with suspected MPN based on patient’s history, clinical examination and blood results. In our study, four novel CALR mutations were discovered that have yet not been presented in the literature. Before referral to the Department of Haematology patients should be screened for the most common causes of thrombocytosis. Based on our results iron deficiency is most commonly unrecognized as it can be present even in the absence of anaemia. Triple negative patients with suspected MPN should be considered as patients with ET as their thrombotic risk can be compared to the risk of CALR-positive patients. These patients should be treated at least with acetylsalicylic acid based on their individual risk for thrombotic complications. SBA is not as efficient in diagnosing ET as a haematologist