Autoimmune Lymphoproliferative Syndrome in Children with Nonmalignant Organomegaly, Chronic Immune Cytopenia, and Newly Diagnosed Lymphoma

This study investigated the frequency of and predictive factors for autoimmune lymphoproliferative syndrome (ALPS) in children with lymphoma, chronic immune cytopenia, and nonmalignant organomegaly. Thirty-four children with suspected ALPS (n=13, lymphoma; n=12, immune cytopenia; n=9, nonmalignant organomegaly) were included. Double-negative T-cells, lymphocyte apoptosis, and genetic findings were analyzed. Patients were stratified into two groups as proven/probable ALPS and clinically suspected patients according to the ALPS diagnostic criteria. Of the 34 patients, 18 (53%) were diagnosed with proven/probable ALPS. One patient had a mutation (c.652-2A>C) in the FAS gene. The remaining 16 (47%) patients were defined as clinically suspected patients. Predictive factors for ALPS were anemia and thrombocytopenia in patients with lymphoma, splenomegaly and lymphadenopathy in patients with immune cytopenia, and young age in patients with nonmalignant organomegaly. ALPS may not be rare in certain risk groups. Our study indicates that screening for ALPS may be useful in children having lymphoma with cytopenia at diagnosis, in those having nonmalignant organomegaly with immune cytopenia, and in those having chronic immune thrombocytopenic purpura or autoimmune hemolytic anemia with organomegaly developing during follow-up.

This study investigated the frequency of and predictive factors for autoimmune lymphoproliferative syndrome (ALPS) in children with lymphoma, chronic immune cytopenia, and nonmalignant organomegaly. Thirty-four children with suspected ALPS (n=13, lymphoma; n=12, immune cytopenia; n=9, nonmalignant organomegaly) were included. Double-negative T-cells, lymphocyte apoptosis, and genetic findings were analyzed. Patients were stratified into two groups as proven/probable ALPS and clinically suspected patients according to the ALPS diagnostic criteria. Of the 34 patients, 18 (53%) were diagnosed with proven/probable ALPS. One patient had a mutation (c.652-2A>C) in the FAS gene. The remaining 16 (47%) patients were defined as clinically suspected patients. Predictive factors for ALPS were anemia and thrombocytopenia in patients with lymphoma, splenomegaly and lymphadenopathy in patients with immune cytopenia, and young age in patients with nonmalignant organomegaly. ALPS may not be rare in certain risk groups. Our study indicates that screening for ALPS may be useful in children having lymphoma with cytopenia at diagnosis, in those having nonmalignant organomegaly with immune cytopenia, and in those having chronic immune thrombocytopenic purpura or autoimmune hemolytic anemia with organomegaly developing during follow-up.
The aim of the present study was to investigate the frequency and predictive factors of ALPS in children with recently diagnosed lymphoma, chronic nonmalignant organomegaly, and chronic immune cytopenia.

Materials and Methods
In total, 34 consecutive patients were included in this study with a two-stage cross-sectional design: those with nonmalignant organomegaly, chronic immune thrombocytopenic purpura (cITP), or autoimmune hemolytic anemia (AIHA) (n=21) between March 2011 and April 2013, and those with newly diagnosed lymphoma (n=13) between June 2013 and March 2015. Patients were also stratified into two groups as proven/probable ALPS (Group 1, n=18) and clinically suspected patients (Group 2, n=16) according to the ALPS diagnostic criteria [17] (Figure 1). The institutional review board approved the study.

Results
The demographic data for ALPS are summarized in Table 1. Of the 34 patients enrolled, 18 (53%) fulfilled the diagnostic criteria for proven ALPS (n=13; 38%) or probable ALPS (n=5; 15%) in Group 1. The remaining 16 (47%) were clinically suspected patients in Group 2. There were significant differences in terms of age between Group 1 and Group 2 (p<0.05). The median age of the patients with nonmalignant organomegaly in Group 1 was significantly lower than that of the nonmalignant organomegaly patients in Group 2 (3 vs. 10 years; p<0.05). The proportions of patients with splenomegaly and lymphadenopathy were significantly higher among the cITP and AIHA subgroups in Group 1 than among the cITP and AIHA subgroups in Group 2 (p<0.05). The proportions of patients with anemia and thrombocytopenia were significantly higher among the lymphoma subgroups in Group 1 than among the lymphoma subgroups in Group 2 (p<0.05).
All relevant data of the 18 patients with proven and probable ALPS are given in Table 2. Of them, 7 (38%) had lymphoma, 5 (28%) had nonmalignant organomegaly, 4 (22%) had cITP, and 2 (12%) had AIHA. Of the seven children with lymphoma, histopathological examination revealed five with Hodgkin lymphoma. Only two of them were positive for Epstein-Barr virus (EBV). Heterozygous splicing mutation in the FAS gene (c.652-2A>C in intron 7) was identified in Case 10 as shown in Table 2. The FAS mutation rate was found to be 20% among patients with nonmalignant organomegaly (n=5).
Five of the 18 children in Group 1 had been scheduled for splenectomy for massive splenomegaly. Splenectomy was canceled after the diagnosis of ALPS. Three of them responded to steroids and mycophenolate mofetil (MMF), one was unresponsive to steroids and MMF but responded to sirolimus, and one received an allogeneic stem cell transplantation. The remaining seven patients with lymphoma received chemotherapy. Four patients with cITP received mostly on-demand treatment with either steroids or IVIG. Two patients with AIHA received steroids and rituximab, which initially controlled the anemia. MMF was given to both patients who were diagnosed with cITP and AIHA (Cases 13 and 17 in Table 2).

Discussion
Patients with ALPS have heterogeneous phenotypes that can mimic malignancy and infectious or autoimmune diseases. Long-term follow-up studies demonstrated ALPS mutation in 15% and 85% of involved subjects [3,7,8,9,10]. In this study, proven or probable ALPS was recorded in 53% of suspected patients. However, the FAS mutation rate was found to be 20% among patients with nonmalignant organomegaly.
Lymphadenopathy and splenomegaly are the most common clinical signs of ALPS, as described in our study [19]. Most patients with type Ia develop lymphoproliferation at a median age of 1.8 years [20]. The same clinical pattern was also described incidentally in a 1-year-old girl with FAS mutation in our study. However, the median age at presentation was 4.9 years in patients with undefined ALPS type III [20]. Accordingly, we found the median diagnostic age as 3 years in undefined ALPS patients Autoimmunity is the second most common sign with the highest probability of requiring medical intervention. The frequencies of ALPS in the subjects with chronic immune cytopenia and in patients with Evans syndrome were 37% and 45%, respectively [7,8,9,10,11,12,13,19]. Similarly, we found ALPS in 34% of patients in Group 1.
Children with immune cytopenia with the occurrence of lymphadenopathy/ splenomegaly during follow-up were approximately 4-to 7-fold more likely to develop ALPS. These findings indicate that lymphadenopathy/splenomegaly may not appear simultaneously in patients with chronic immune cytopenia. Furthermore, positive Coombs tests and hypergammaglobulinemia are frequently observed in patients with Evans syndrome [11,12,13]. We observed that nearly half of the ALPS patients had hypergammaglobulinemia and positive Coombs tests. Development of lymphadenopathy, splenomegaly, and autoantibodies during follow-up in children with cITP and AIHA should alert the physician to a possible diagnosis of ALPS.
Lymphoma is usually diagnosed in patients with ALPS at advanced ages. Lymphoma at a median age of 17 years in both adults and children with ALPS was reported in one study [8]. However, the median age of lymphoma patients was found as 12 years in our study. Most reported patients with ALPS have Hodgkin lymphoma, and EBV is positive in these cases [15,16]. Similarly, our patients were mostly diagnosed with Hodgkin lymphoma, but investigations of these patients revealed only two cases with EBV. In addition, children with newly diagnosed lymphoma with the presence of anemia and thrombocytopenia were approximately 3-to 4-fold more likely to develop ALPS. Our study indicated that the presence of anemia and thrombocytopenia in patients with lymphoma at diagnosis may be useful for ALPS screening.
Splenectomy and rituximab are not recommended in ALPS because of sepsis and recurrence risk in most cases [1,2,3,21,22,23,24]. We canceled the scheduled splenectomies for five patients with massive splenomegaly. Furthermore, some patients with cITP and AIHA might be resistant to standard treatment, as in previous reports [25,26]. Partial response to rituximab was observed in cases of AIHA. We believe that treatment response could help the physician reach a possible diagnosis of ALPS in children with cITP, AIHA, and nonmalignant organomegaly. The major limitations of the present study were that the other ALPS-related genes [27,28,29,30] were not studied due to lack of resources and all lymphoma cases/adult cases were not included.
Our data indicate that investigation of ALPS is warranted in children with lymphoma presenting with cytopenia, in cases of chronic nonmalignant organomegaly with immune cytopenia, and probably in patients with cITP and AIHA developing organomegaly during follow-up.