Introduction

Immunotherapies are now recognized treatments in the fight against cancer. They are innovative therapies that have revolutionized the established prognosis of many types of cancer. However, some information is lacking, particularly concerning their adverse effects, which can be life threatening. The latter are on the rise due to the increase in the number of indications for immunotherapies. It therefore seems essential to identify them in order to prevent them. One of the adverse effects of immunotherapy that can be life threatening is haemophagocytic lymphohistiocytosis (HLH) [1-4], which is difficult to diagnose because of its non-specific symptoms. In our article, we report our experience with HLH during the management of metastatic ovarian cancer with a BOLD (Bevacizumab, Olaparib, Durvalumab) combination therapy. Currently, this is the only existing report of HLH under immunotherapy but also with EBV reactivation. We can thus ask ourselves the question of what is the influence of EBV on the pathogenesis of this HLH and what practical therapeutic consequence it can generate.

Case Presentation

We present the case of a 60-year-old female patient followed in oncology for a recurrent ovarian cancer. She had already received several lines of treatment and was in relapse (lymph node metastases and peritoneal carcinosis). She was then included in a protocol called BOLD (ClinicalTrials.gov Identifier: NCT04015739). It is a trial evaluating the safety and efficacy of the combination of bevacizumab, Olaparib and Durvalumab (MEDI 4736) in patients with high-grade serous or endometrioid ovarian tumour or other high-grade non-mucinous epithelial tumour, who have received at least one prior platinum-taxane-based chemotherapy regimen, and who have platinum-resistant disease (PRR) or platinumsensitive relapse (PSR), regardless of the line of chemotherapy administered at the time of relapse. The patient received a first course of BOLD but presented 6 weeks later (fifth course) a poor clinical tolerance with the appearance of a fever of 39° and dyspnoea at the slightest effort. Cardiopulmonary auscultations did not reveal any abnormality. On palpation, we found splenomegaly with a known left supra-clavicular adenopathy of tumoral origin. Biologically, we found pancytopenia (haemoglobin 6.7 g/L, leukocytes 1.26 G/L including neutrophils 0.66 G/L, platelets 12 G/L) with a CRP 174 mg/L. The blood smear did not show any schizocytes but the haptoglobin was collapsed with LDH at 1440). The liver workup is altered (AST 219, ALT 183, GGT, 709, PAL 384, total bilirubin 26). Serum ferritin was measured at 28 g/L. Imaging (CT scan) and bacteriological examinations (blood cultures + urinary cytobacteriological examination) were negative. Given this clinical and biological picture, the diagnosis of HLH was evoked. The patient’s poor clinical tolerance led to the suspension of immunotherapy and platelet and blood transfusions. Concerning the etiological work-up, two attempts at myelograms were made but were inconclusive, eliminating a leukaemia or myelodysplasia. The various serology’s (Hepatitis B, Hepatitis C, CMV, PVB19 and HHV6) were negative except for an old serology against EBV (identified with positive anti-VCA IgG antibodies but negative anti-VCA IgM) and an EBV viral load was detectable at 4.7 log. Macrophagic activation syndrome under immunotherapy is retained in this patient with EBV viral reactivation as a favouring factor. A single course of Etoposide was given with 20 mg of dexamethasone and a dose of 0.4 mg/kg of type G immunoglobulin. The patient’s clinical condition was normalized within 48 hours after initiation of treatment. Her only symptom was dyspnoea for several weeks and she finally returned home with biweekly biological monitoring. At the 30th day followup, the patient was symptom free (except for the persistence of the left supra-clavicular lymph node) and her biological check-up was normalized (notably Hb 11.1 and CRP 7 mg/L).

Discussion

Among the many etiologies of HLH, immunotherapy [4] and viral infection [3-6] are important to know because they are preventable. Moreover, they have a good prognosis when managed in time. In our case, the onset of immuno-induced HLH was concomitant with EBV infection. This article is the first to report a HLH on durvalumab. The prognosis of EBV-HLH was until recently appalling because no antiviral treatment had been shown to be effective. The use of high-dose dexamethasone [2] and etoposide (VP-16) [3] has finally proven to be effective, regardless of the origin of the HLH (immuno- or virus-induced). The estimated survival rate was 59-75%. The main predictive factor of response was administration within 2 days of the highest ferritin concentrations. The majority of case reports published to date concerning immuno-induced HLH confirm a clinical and biological normalization a few weeks after initiation of treatment [1-12]. However, the risk of Etoposide-induced acute myeloid leukaemia remains significant. Some protocols offer the option of cyclosporine as a treatment, but no clear benefit has been demonstrated. It avoids exposure to chemotherapy and the risk of bone marrow aplasia associated with Etoposide. Etoposide avoids the risk of severe reactions that can be associated with cyclosporine, and may be less immunosuppressive to T cells [8]. Rituximab [7] suppresses B cells (which are a reservoir for EBV), whereas Etoposide and Dexamethasone target T and NK cells (and thus promote EBV clearance and elimination [9]). One study concluded that Rituximab was very effective in patients whose B cells were affected by EBV. It is therefore not effective in patients whose EBV virus affects T and NK5 cells. Some studies have shown the therapeutic role of molecules such as Emapalumab or Ruxolitinib in the management of HLH. Their mechanism of action is not cellular but rather against cytokines. It would therefore be useful in future studies to determine whether these latter molecules are sufficiently effective to suppress Etoposide in the management of HLH. In addition, one study has monitored the efficacy of Nivolumab in the treatment of HLH. Nivolumab was found to be effective against all subtypes of cells affected by the virus (T,B,NK). These results suggest the importance of anti-PD1 molecules in the treatment of MAS [10]. A retrospective study [4] identified certain factors that increase the risk of immuno-induced HLH. These are mainly age over 60 years, male gender, and the length of time since the start of immunotherapy. Genetics (in particular the PRF1A91V3 gene polymorphism) also plays a role in the development of HLH. The risk of immunotherapy in the occurrence of HLH being established, the knowledge of additional factors is important in order to recognize the clinical warning signs as soon as possible. Cancer itself is a cause of HLH, in this case called paraneoplastic HLH [11]. The latter remains a rare complication but has a survival rate of < 25% at one year5. The delay in diagnosis is long because this pathology is rare and with a broad and non-specific clinic. Its recognition is a priority in order to initiate adequate treatment as early as possible. In this case, the basis of treatment of HLH is etiological treatment: A study [11] reporting MAS in a patient with ovarian cancer. Complete symptom resolution resulted from tumour resection and medical treatment. Unfortunately, malignancy-triggered HLH is associated with increased mortality. A study [13] present a series of nine paediatric and young adult patients in whom an initial diagnosis of HLH delayed the discovery of underlying malignancy and frequently delayed truly curative therapy. All of these patients ultimately died from multiorgan failure with active malignancy present. Both patients who received full-dose chemotherapy are still alive with no evidence of disease. These cases highlight the importance of distinguishing malignancy-associated HLH in order to limit morbidity and mortality. One paediatric series [14] describing a 6-month overall survival of 67% and median overall survival of 1.2 years, with the majority of deceased patients having active malignancy at the time of death [15].

Conclusion

In this article, we report our experience in the management of immuno- and virus-induced HLH in stage 3B ovarian cancer with Etoposide, Dexamethasone and Immunoglobulin G. New molecular targets appear to offer a therapeutic solution with fewer complications than Etoposide but need to be further investigated. It is important to initiate treatment as soon as possible, especially when the origin of HLH is curable. In other cases, it is necessary to know how to recognize and manage it early because its prognosis remains poor with a low survival rate at 1 year.

References

1. Kalmuk J, Puchalla J, Feng G, Giri A, Kaczmar J (2020) Pembrolizumabinduced Hemophagocytic Lymphohistiocytosis: an immunotherapeutic Cancers Head Neck. 5: 3.
2. Takahashi H, Koiwa T, Fujita A, Suzuki T, Tagashira A, et al (2020)A case of pembrolizumab-induced hemophagocytic lymphohistiocytosis successfully treated with pulse glucocorticoid therapy. Respiratory Medicine Case Reports. 30: 101097.
3. Al-Samkari H, Snyder GD, Nikiforow S, Tolaney SM, Freedman RA, et al (2019) Haemophagocytic lymphohistiocytosis complicating pembrolizumab treatment for metastatic breast cancer in a patient with the PRF1A91V gene polymorphism. J Med Genet. 56: 39-42.
4. Noseda R, Bertoli R, Müller L, Ceschi A (2019) Haemophagocytic lymphohistiocytosis in patients treated with immune checkpoint inhibitors: analysis of WHO global database of individual case safety j immunotherapy cancer. 7: 117.
5. Meng GQ, Wang JS, Wang YN, Wei N, Wang Z (2021) Rituximabcontaining immuno-chemotherapy regimens are effective for the elimination of EBV for EBV-HLH with only and mainly B lymphocytesof EBV infection. International Immunopharmacology. 96: 107606.
6. Akagi Y, Awano N, Inomata M, Kuse N, Tone M, et al (2020) Hemophagocytic Lymphohistiocytosis in a Patient with Rheumatoid Arthritis on Pembrolizumab for Lung Adenocarcinoma. Intern Med. 59:1075-1080.
7. Marsh RA (2018) Epstein Barr Virus and Hemophagocytic Lymphohistiocytosis. Front Immunol. 8: 1902.
8. Gonzalez F, Vincent F, Cohen Y (2009) Syndrome d’activation macrophagique d’origine infectieuse : étiologies et prise en charge. Réanimation. 18: 284-290.
9. Imashuku S, Morimoto A, Ishii E (2021) Virus‐triggered secondary hemophagocytic lymphohistiocytosis. Acta Paediatrica. 110: 2729
10. Liu P, Pan X, Chen C, Niu T, Shuai X, et al (2020) Nivolumab treatment of relapsed/refractory Epstein-Barr virus–associated hemophagocytic lymphohistiocytosis in adults. Blood. 135: 826-833.
11. Nosratian-Baskovic M, Tan B, Folkins A, Chisholm KM, Dorigo O(2016) Hemophagocytic lymphohistiocytosis as a paraneoplastic syndrome associated with ovarian dysgerminoma. Gynecologic Oncology Reports. 17: 38-41.
12. Sadaat M, Jang S (2018) Hemophagocytic lymphohistiocytosis with immunotherapy: brief review and case report. Journal for ImmunoTherapy of Cancer. 6: 49.
13. Gurunathan A, Boucher AA, Mark M, Prus KM, O’Brien MM, et al (2018) Limitations of HLH-2004 criteria in distinguishing malignancyassociated hemophagocytic lymphohistiocytosis. Pediatric Blood & 2018: e27400.
14. Lehmberg K, Sprekels B, Nichols KE, Woessmann W, Muller I, et al(2015) Malignancy-associated haemophagocytic lymphohistiocytosis in children and adolescents. Br J Haematol. 170: 539-549.
15. Ramos-Casals M, Brito-Zerón P, López-Guillermo A, Khamashta MA, Bosch X (2013) Adult haemophagocytic syndrome. The Lancet. 383:1503-1516.