We showed that the prevalence of chromosome 8p11.2 abnormality in Japanese clinical practice was 0.06%, which is similar to that reported by Mayo Hospital of 0.06%; suggesting that there is no ethnical difference in the frequency of 8p11.2 abnormalities in hematological diseases [5]. In addition, not all patients with 8p11.2 abnormalities were diagnosed with MLN-FGFR1 abnormalities. About one third of them were categorized as MLN-FGFR1 abnormalities in our study, which was also consistent with a previous report in which four of 14 patients with 8p11.2 translocations and in another report, four of 12 patients with 8p11.2 translocations were diagnosed as MLN-FGFR1 abnormalities [5, 6].

Although the prevalence of chromosome 8p11.2 abnormalities in Japan was similar to that in Western countries, the disease types were different. The majority of Japanese patients with 8p11.2 abnormalities had lymphoid malignancies, whereas those in Western countries had myeloid malignancies. However, we cannot find a reasonable explanation for this discrepancy.

All three cases with MLN-FGFR1 abnormalities in this report harbored t(8;13)(p11.2;q12), which is the most common chromosomal abnormality in MLN-FGFR1 abnormalities [1]. In this translocation, the partner gene with FGFR1 is ZNF198 (previously reported as ZMYM2, FIM, and RAMP) located at 13q11-12, and the zinc finger domain of ZNF198 is fused to the tyrosine kinase domain of FGFR1. Recipient mice transplanted with bone marrow cells transfected with ZMYM2::FGFR1 developed myeloproliferation and intestinal T-cell lymphoma, and in immunodeficient mice, human cord blood CD34+ cells transfected with ZMYM2::FGFR1 showed expansion of multiple myeloid cell lineages and accumulation of blasts in BM [7, 8]. In patients with t(8;13)(p11.2;q12), lymphadenopathy and hepatosplenomegaly are often the first symptoms, and most patients are diagnosed with T-LBL/T-lymphoma [1]. Consistent with this, all three patients in our study developed T-cell malignancies in their clinical course.

Umino et al. reported the characteristics and prognosis of 45 cases with MLN-FGFR1 abnormalities collected from a computerized search of the medical literature using PubMed® [9]. At diagnosis, 31% and 69% of patients were in chronic phase as MPN or MDS and in advanced phase as acute leukemia or lymphoblastic lymphoma, respectively. Approximately half of patients in chronic phase transformed to advanced blast phase within one year, and the median OS from diagnosis was 9 months if allogeneic HSCT was not performed before transformation to advanced blast phase. The OS for patients in advanced phase was dismal, and the 1-year OS was 29.8%. Since FGFR1 rearrangement leads to constitutive activation of tyrosine kinase, which induces abnormal proliferation, inhibitors targeting FGFR1 kinase activity have been developed [10, 11]. Among them, pemigatinib, a reversible ATP-competitive FGFR inhibitor, has been recently approved as a novel drug for treatment of myeloid/lymphoid neoplasms with FGFR1 rearrangement in the US and Japan. To select appropriate patients for pemigatinib therapy, FISH analysis demonstrating FGFR1 gene disruption would be desirable in addition to chromosomal 8p11.2 abnormalities for patients who do not exhibit the typical clinical features of MLN-FGFR1 abnormalities, because approximately one third of patients with 8p11.2 abnormalities were diagnosed as MLN-FGFR1 abnormalities in our study and previous reports [5, 6].

Of other seven patients without typical features of MLN with FGFR1 rearrangement, one patient was diagnosed as AML M5b with t(8;16)(p11.2;p13.3). MYST3 (MYST histone acetyltransferase 3) is disrupted by the chromosomal 8p11.2 translocation, as is FGFR1, and the most common translocation partner gene for MYST3 is CEBBP (CREB-binding protein), located on chromosomal 16p13.3 [12]. As the chromosomal translocation t(8;16)(p11.2;p13.3) is reported to be associated with an aggressive form of AML M4/M5, the 8p11.2 abnormality detected in this patient is speculated to be a MYST3-CREBBP rearrangement, although we cannot show it by FISH analysis due to lack of residual BM samples [13]. Another patient whose MDS progressed to AML had FGFR1 amplification. The overexpression or amplification of FGFR1 has been reported in 10–20% of breast cancer, head and neck squamous cell carcinoma, and lung squamous cell carcinoma, and its association with poor survival has been reported in breast cancer, head and neck squamous cell carcinoma, but not in lung cancer [14,15,16]. In contrast, overexpression or amplification of FGFR1 was rare in hematological malignancies. For the remaining four patients (patients 1, 7, 8, and 10), FISH examination using dual-color split probes for the 8p11.2 loci was not performed because of the lack of residual samples; therefore, it is difficult to completely exclude them from having MLN with FGFR1 rearrangement. The breakpoint of partner genes rearranged with 8p11.2, in these four cases included 1p13, 1q12, 8p11.2, and 15q11.2. As these chromosomal breakpoints have not been previously reported to generate FGFR1 fusion genes and cause MLN with FGFR1 rearrangement [1], the possibility of their diagnosis as MLN with FGFR1 rearrangement might be low, along with the absence of the typical clinicopathological features of MLN with FGFR1 rearrangement.

In conclusion, the prevalence of 8p11.2 abnormalities in Japanese clinical practice and the proportion of MLN-FGFR1 abnormalities among them were almost consistent with those in previous reports from Western countries.