Classical Philadelphia chromosome-negative (Ph-negative) myeloproliferative neoplasms (MPNs) include the clinical entities of polycythaemia vera (PV), essential thrombocythaemia (ET) and primary myelofibrosis (PMF). MPNs are associated with a significant reduction in patients’ quality of life and an increased mortality risk [1
]. Significant efforts are ongoing to develop therapies that halt MPN progression; epidemiological data are crucial to guide such efforts.
MPNs are not uncommon. A systematic review reported the combined international incidence rates to be 0.84, 1.03, and 0.47 per 100,000 for PV, ET, and PMF, respectively [4
]. There was wide variation in incidence rates in the reviewed studies, which included global data from 1935 to 2010. Newer data from the Surveillance, Epidemiology and End Results (SEER) Program in the USA reported overall incidence rates from 2001 to 2016 to be 1.18, 1.14, and 0.33 per 100,000 for PV, ET, and PMF, respectively [5
]. The Australian Cancer Registry started collecting MPN data in 2003 and reported incidence rates between then and 2014 to be 0.90, 0.95, and 0.45 per 100,000 for PV, ET, and PMF, respectively [6
Internationally, differences in gender, age and ethnicity have been reported to affect the outcomes of MPN. Gender incidence rates are higher in males for PV and PMF, and higher in females for ET [4
]. Incidence and mortality risk increase with age; the median age of diagnosis is in the 60–70 age range in nations with mostly European ethnic populations [5
], and in the 50–60 age range in Asian nations [11
]. The excess hazard was near 3.0 (95% CI: 2–4) for those aged 50–69 compared to those aged 15–49, and near 7.3 (6–10) in those aged 70–89 [5
MPN have different incidence rates across ethnicities in the USA [8
]. Whites have the highest incidence of PV (incidence rate (IR)—12.3) and PMF (IR—3.3) compared to Hispanics (PV IR—7.2; PMF IR—2.2), Blacks (PV IR—7.5; PMF IR—2.4), Asian and Pacific Islanders (PV IR—7.5; PMF IR—2.4). For ET, Blacks have the highest incidence (IR—11.5) compared to other ethnic groups [8
]. Black patients diagnosed with PV (Hazard ratio (HR): 1.51; p
= 0.004) and ET (HR: 1.63; p
= 0.001) have a higher risk of death compared to White patients [15
]. The authors suggested this may have been due to differential access to healthcare. In a single institution study where all patients had equal access to healthcare, Black patients with MPN were at higher risk of haemorrhagic complications than White patients, but there was no difference in overall survival [16
Ethnic disparities have also been observed for PV in New Zealand [17
]. In a single centre study, New Zealand (NZ) Polynesian patients were diagnosed with PV at a significantly younger age (mean 54 years) than Europeans (mean 68 years), but there was no difference in overall survival. Nevertheless, due to the younger age at diagnosis, NZ Polynesian patients had shorter life expectancies compared with the general population [17
]. There are no other epidemiological or outcome data for MPN in New Zealand.
The New Zealand Cancer Registry (NZCR) was established in 1984 and has been collecting data for PV and ET from 2010 and for PMF from 2014. The aim of this study was to review the demographics and outcomes of patients with PV, ET and PMF in New Zealand and, secondly, to describe the capture rates of these MPNs by the NZCR. The overriding aim was to assist further clinical research on MPN in New Zealand. Findings are expected to identify at risk populations and such epidemiological data may guide MPN screening recommendations in New Zealand and globally.
This study investigated the incidence and mortality outcomes of patients with classical Ph-negative MPN (PV, ET and PMF) in New Zealand based on the data reported to the NZCR between 2010 and 2017 for PV and ET, and between 2014 and 2017 for PMF. We found that capture rates of MPN increased in 2014 and peaked in 2017, reaching annual incidence rates of 3.38, 0.90, 1.56 and 0.92 per 100,000 for all MPNs, PV, ET, and PMF, respectively. Māori and Pacific Island patients presented at younger ages in PV, ET and PMF and died earlier in PV and PMF. Male gender was a risk factor for mortality in PV and PMF, with nearly double the risk of mortality. Māori ethnicity nearly tripled the risk of mortality in PMF. There were no significant demographic risk factors for mortality identified for patients with ET. Our study provides important and novel epidemiological findings not previously known about MPNs in New Zealand.
In 2016, the NZCR reporting guidelines were introduced and in effect, MPN registrations peaked in 2017. The biggest increase was for PV (from 0.60 per 100,000 in 2016 to 0.90 in 2017), suggesting that prior to 2017, many PV cases were not reported to the NZCR if diagnosed solely on the peripheral blood and molecular criteria. Overall, average MPN capture rates in New Zealand between 2010 and 2017 (0.76, 0.99, and 0.82 per 100,000 for PV, ET, and PMF, respectively) are within previously reported international incidence ranges of 0.01–2.61, 0.21–2.27, and 0.22–0.99 per 100,000 for PV, ET, and PMF, respectively [4
]. However, they are lower than the average international rates for PV and ET (0.84 and 1.03 per 100,000, respectively), but higher than the average international incidence rate for PMF (0.47 per 100,000) [4
]. PMF rates captured by the NZCR are of the highest in the world (0.82 per 100,000 compared with 0.3–0.4 from 2001 to 2016 in the USA and ~0.45 between 2003 and 2014 in Australia) [5
]. It is possible that some patients had secondary myelofibrosis (MF), either post-PV or -ET. It can be challenging to diagnostically differentiate secondary from primary MF, as well as ET from early PMF [20
]. If MF was diagnosed late in disease progression, an antecedent PV or ET may not have been captured. Underreporting and late reporting may have introduced bias; follow-up analyses of the NZCR data should help clarify our findings.
The median age of MPN patients diagnosed in nations with majority European ethnic populations is generally in the 60s to early 70s, with the youngest being for ET and oldest for PMF [5
]. This was similar in our study—mean age at diagnosis was 65.5 years for ET patients and 69.2 years for PMF patients. Gender differences between NZCR and international data were also similar, showing higher incidence among females for ET and males for PV and PMF [4
Due to short-term follow-up, our mortality rates are difficult to compare to USA cohorts, but there are similarities. PMF is known to be associated with younger age of death, and ET has good survival rates [1
]. This is reflected in the best outcomes and older age at death in ET (mean 80.2 years), and younger age at death in PMF (mean 74.8 years) in our cohort. However, the comparable survival between PMF and PV in our study is unexpected. This is likely due to late reporting of PV cases (suggested by their relatively older age and low PV capture rates), further confounded by the short time period of PMF records available (2014–2017). Despite these limitations, we identified male gender as an independent risk factor for mortality in PV and PMF, consistent with other reports [23
]. Recent study demonstrated that men have a higher mutational burden in CD34-positive cells and higher risk of non-MPN-specific high-risk mutations including in ASXL1
, and IDH1/2
genes compared with women [25
]. It would be interesting to test for such mutations in New Zealand patients in the future.
Our work uncovered some intriguing ethnic differences in demographic characteristics of MPN in New Zealand. For PV, Māori and Pacific Islanders were younger at diagnosis than Europeans (median age of 63.0, 56.0 and 73.0 years, respectively). Māori and Pacific Islander patients also died younger than European patients (median age of 68.7, 66.1 and 81.0 years, respectively). We have previously reported that NZ Polynesians with PV treated at Middlemore Hospital (Auckland) presented at a significantly younger age than Europeans [17
], and that finding has been replicated in this study.
For ET, the median age of diagnosis in Māori and Pacific Islanders was also significantly younger than Europeans, but not as pronounced as for PV patients. However, the median age of death from ET was not significantly different between the ethnic groups, suggesting the natural history of ET is similar between ethnicities.
The median age of diagnosis of PMF for Māori and Pacific Islanders was significantly lower than Europeans (57.0, 62.5 and 71.0 years, respectively). Māori PMF patients also died at a younger median age than Pacific Island and European PMF patients (60.3, 75.9 and 75.4 years, respectively). The percentage of Māori PMF patients that died was higher than Pacific Island and European PMF patients (60.0%, 40.0% and 46.6%, respectively). There was an unusually high percentage (80%) of Māori and Pacific Islanders males with PMF, but neither of these percentages reached significance. After adjusting for age, gender and ethnicity, Māori ethnicity tripled the risk of PMF mortality compared to Europeans (HR: 2.94, 95% CI: 1.37–6.3; p = 0.006).
The Dynamic International Prognostic Scoring System-Plus (DIPSS-Plus) uses a set of risk factors to predict survival in PMF [26
]. It is unknown whether Māori ethnicity increases the likelihood of having these risk factors, and further studies are required. Cardiovascular complications are a common cause of death in PMF [27
]. Māori have higher rates of cardiovascular disease than Europeans and Pacific Islanders [28
], which may have contributed to higher death rates. Māori are also overrepresented in areas of higher deprivation; thus societal inequities in access to healthcare may have affected worse outcomes [29
Our previous research found that NZ Polynesian patients with PV [17
] and acute promyelocytic leukaemia [30
] also present at a younger age. Together, this raises a possibility that NZ Polynesians have a unique set of risk factors or a form of genetic predisposition to myeloid neoplasia. Polynesians smoke more tobacco and have higher rates of obesity, which increases the risk of certain cancers [31
], and is of relevance in MPN [34
]. Risk factors for MPN remain poorly defined; however, certain modifiable lifestyle factors have been implicated (e.g., diet, smoking, alcohol and coffee consumption, psychological stress and physical activity), as well as certain chemical exposures (e.g., benzene and polycyclic aromatic hydrocarbons) [39
]. Better definition of the roles of these risk factors in MPN pathogenesis may help design new preventative strategies against MPN. Future studies in New Zealand should therefore record lifestyle and environmental risk factors in the MPN cohorts to determine how these factors impact MPN presentation and outcomes in Polynesian and European patients. Evaluation of associated inflammation would also be of interest, in particular in the bone marrow microenvironment [41
Inherited factors play a role in the development and biology of MPN [42
]. Research into genomic associations between ethnicity and MPN incidence and outcomes in the New Zealand population is important to undertake. Similar genetic associations, for example the E-Cadherin (CDH1
) gene mutations of hereditary diffuse gastric cancer and its familial associations with Māori ethnicity, have had a revolutionary clinical impact in the New Zealand setting [44
]. We advocate for genome-wide association studies to be performed in NZ Polynesian and European patients with and without MPN to seek an underlying cause. Results may ultimately lead to novel preventative, diagnostic and therapeutic measures for NZ Polynesians and other patients.
More research into ethnic differences in haematological cancers is needed globally. Further advances in clinical outcomes will require an understanding of the drivers of ethnicity-based differences in cancer outcomes. A better understanding of these drivers will allow prioritisation of research and clinical funding to serve populations with more need [45
]. Future discoveries in this area have the potential to improve both patient and public health outcomes. The results may have implications for recommending increased screening of extended families of MPN patients.
Our study should be considered in light of its limitations. PMF data were available only from 2014 to 2017, covariates were limited, and the follow-up time was short compared to other larger cohorts [5
]. Many confounders such as comorbidities and patient factors influence patient mortality; however, these were not available for analysis and should be considered in the design of future clinical studies. Clinicopathological variables, in particular indicators of disease severity, would be important confounders to adjust for in future analyses of MPN mortality. Other studies have adjusted for laboratory variables such as white blood cell counts, cytogenetics and molecular results, splenomegaly, thrombotic events, major bleeding events, and transplant status [16
]. The reason why risk factors for ET mortality were not identified is unclear but some of the confounding factors above and inadequate reporting may have contributed. The NZCR may not yet comprehensively captured the true incidence of reported MPNs, but this is a continually improving process. Further reviews of the NZCR data will be undertaken in the future to help clarify true incidence trends and risk factors for MPN in New Zealand.
A strength of this study is its relatively large population size for each MPN subgroup. As this cohort includes patients from all over New Zealand, results are likely generalisable to other populations and provide insights about MPN that may be useful in an international context. Ruxolitinib, prescribed to improve MPN signs and symptoms, has only been funded in New Zealand since October 2018. Therefore, the NZCR outcome data are not impacted by the use of this drug. As novel therapies are introduced moving forward, NZCR provides an invaluable source of unbiased information to monitor impact of new treatments on patient survival. Our results highlight groups of MPN patients that are at high risk of mortality in New Zealand that should be reached with novel therapies—these are PMF patients, in particular Māori and male patients. Baring limitation of biased reporting, PV patients also did not fare well, highlighting the need for novel treatments for high-risk PV patients.