Mucositis has become the dose-limiting complication of high-dose chemotherapy. Therefore, recombinant keratinocyte growth factor (KGF)-1, Palifermin, seems a major breakthrough in the management of patients receiving intensive treatment for solid tumours and haematological malignancies.1 Palifermin reduces the incidence and severity of oral mucositis and its clinical consequences.1 The side effects of treatment are generally mild and transient, and are confined to skin rashes, pruritus, mouth and tongue disorders and altered taste.1 This led the Food and Drug Administration (FDA) to approve palifermin for the prevention of oral mucositis of patients with haematological malignancies and boosted the development of other mucosa-protective growth factors, such as repifermin (KGF-2) and velafermin (fibroblast growth factor-20).
Preclinical studies show that KGF has several other properties besides protecting epithelial surfaces, as it lessens apoptosis of several cell types and modulates the inflammatory and immune responses,2 which could be of benefit in the setting of allogeneic haematopoietic SCT. Moreover, KGF has been shown to reduce the incidence of acute GVHD (aGVHD) in mouse models3 and to improve immune reconstitution by thymic preservation.4 The amelioration of aGVHD has been attributed to the preservation of gut integrity with decreased translocation of microbial products and reduced cytokine production, although other factors probably contribute such as the shift of cytokine profiles from helper T cell (Th) 1 to Th2.
Therefore, with palifermin proven to reduce mucositis in humans treated with intensive chemotherapy and radiotherapy, it is conceivable that the drug should lead to less aGVHD after allogeneic haematopoietic SCT. This question has been addressed recently in three clinical trials of matched-related and -unrelated allogeneic haematopoietic SCT after myeloablative conditioning using CY plus TBI (CY-TBI) and BU plus CY (BU-CY).5, 6, 7 Palifermin was mostly administered at a dose of 60 μg/kg intravenously 3 days before and 3 days after conditioning. The studies were small and consisted of heterogeneous groups of patients and only one was a randomized double-blind trial. None showed an impact on the occurrence of aGVHD, despite the amelioration of oral mucositis. However, the impact on mucositis was not consistent and seemed present only in subgroups, with one study showing an impact on matched-unrelated transplant recipients and another showing an effect on patients conditioned with CY-TBI and not BU-CY.
The addition of MTX (15 mg/m2 on day +1, 10 mg/m2 on days +3, 6 and 11) to the regimen of CsA or tacrolimus for the prophylaxis of GVHD might explain this. Importantly, MTX is cytotoxic to epithelial cells, and aGVHD prophylaxis regimens containing the drug are associated with significantly more moderate and severe mucositis.8 The protective effects of palifermin on epithelial barriers of gut and mouth might, therefore, have been offset by MTX, resulting in less impact on mucositis and failure to reduce aGVHD. Indeed, in a mouse model, MTX-induced mucositis of the small intestine was not reduced by pretreatment with KGF, suggesting that MTX counteracts the effects of KGF.9
In addition to its protective effect on epithelial barriers, KGF also stimulates thymic epithelial cell proliferation and function.4 When given to mice ahead of myeloablative conditioning, KGF reduced thymic epithelial cell injury resulting from myeloablative conditioning and thymic GVHD resulting from alloimmune responses directed at the thymus.4 Thymus preservation results in accelerated post transplant immune recovery predominantly of the T-lymphocyte compartment. Again, MTX could nullify KGF's beneficial effects on the toxicity of chemotherapy on epithelial cells of the thymus. Although little is known about the effects of MTX on the thymus, a study in mice indeed showed MTX to reduce thymus mass even at low dose, compared with calcineurin inhibitor tacrolimus.10 In addition, folic acid is required for normal thymus development with drugs such as MTX interrupting this pathway, resulting in thymic atrophy. Importantly, CsA also negatively influences thymic regeneration and subsequent immune reconstitution, highlighting the need to consider the effects of standard GVHD prophylaxis on the expected protective effects of KGF. Although no clear differences exist in the immune reconstitution after exposure to CsA and MTX, combining the two drugs results in more toxic complications than are seen with either alone.8
Also the timing of palifermin is important for both the intestinal and thymic epithelial preservation and function. KGF was used before, during and after conditioning in most mouse models showing a beneficial effect on aGvHD. This contrasts with clinical trials in humans, as palifermin was not administered in this way, because concomitant use of the drug can paradoxically increase mucositis by increasing the susceptibility of proliferative epithelial cells to cytostatic drugs. The profile of KGF suggests administering the drug later on, either during mucosal and thymic recovery or at the time of active aGvHD. However, studies on the optimal dosing schedule for palifermin are absolutely necessary before firm conclusions can be drawn. In these studies, because of the clear negative effects on epithelial cells, MTX should be avoided altogether to determine the real benefits of palifermin regarding aGVHD and immune reconstitution.