Impairment of lung diffusion capacity—a new consequence in the long-term childhood leukaemia survivors

The survival rate of patients with childhood neoplasm has considerably improved in the last decade [1]. Currently, the number of childhood cancer survivors in Europe ranges between 300,000 and 500,000 [2]. However, survivors are known to be at risk of serious vital organ system toxicities and long-term side effects associated with anti-neoplastic therapy which influence their quality of life [3‐5]. An American multi-institutional study reported that over 60% of cancer survivors after oncology therapy in childhood had at least one, and 20% had multiple chronic health conditions [4, 6]. Observations from Europe are similar and show that more than 50% of the survivors present at least one side effect, with a growing prevalence and severity with time after the completion of oncology treatment [2, 7‐10]. The nationwide study performed in the largest cohort of Polish childhood cancer survivors showed that over 80% of them had one or more symptoms of organ dysfunction [11].

Haematologic malignancies are the most frequent type of childhood neoplasms with the highest prevalence of acute lymphoblastic leukaemia (ALL; 70–80% of patients) and acute non-lymphoblastic leukaemia (ANLL; approximately 10–20% of patients). The treatment of choice for leukaemia is chemotherapy which can be used in conjunction with other therapies in selected cases, such as radiotherapy of the central nervous system, or haematopoietic blood steam transplantation (HSCT). The identification of the individual factors responsible for the impairment of pulmonary function in the treatment of leukaemia is difficult because of the interaction of many variables (cytostatic drugs, radiotherapy, HSCT, pulmonary infections, individual predisposition). Pulmonary toxicity after oncology treatment may become clinically evident many years later, and may adversely affect the long-term survival as well as the quality of life of cancer childhood survivors. Damage to the alveo-capillary barrier and pulmonary fibrosis are well-described as the consequences of the late toxicity in adults [12, 13]. However, much less is known about the delayed pulmonary effects of chemotherapy, and radiotherapy in children. A further important issue is whether and what is the clinical manifestation of late pulmonotoxicity after oncological treatment in childhood. The available data on risk factors for the late adverse events and clinical manifestations of pulmonary dysfunctions in long-term childhood leukaemia survivors (CLS) are conflicting [14‐16].

The aims of this study were (1) to assess pulmonary function in leukaemic survivors after oncology treatment in childhood, (2) to identify treatment-related risk factors, and (3) to identify clinical manifestation of late occurrence impairment of air flow and lung diffusion in CLS.

The study provides insight on the late pulmonary outcomes measured by PFT in childhood leukaemia survivors at least 5 years after completion of oncology treatment (chemotherapy alone or followed by HSCT with additional conditioning megachemotherapy or total body irradiation).

The most important finding of this study is the fact that there is a prolonged difference in the pulmonary function test between survivors who had more aggressive treatment (with HSCT), as well as those who experienced pneumonia, especially of CMV origin, and both survivors without these risk factors and subjects from CG. At least 5 years after the completion of oncology therapy, deteriorated DLCO was demonstrated in most of CLS.

It is well-known that CMV pneumonia in immunocompromised patients is an acute, life-threatening illness with a high mortality rate in spite of intensive treatment with antiviral agents; however, there is little data about the long-term sequelae in such cases. Chien et al. describe seven BMT recipients, after CMV pneumonia. Five patients died, in the remaining two restrictive lung disease with lower DLCO was diagnosed [25]. Also, in kidney transplant recipients, CMV caused subclinical pulmonary dysfunction with pulmonary diffusion disturbances some months after active disease treatment [26]. In our study, children with a CMV infection had significantly lower DLCO, which was particularly expressed in the group with HSCT. In some other studies, CMV-positive leukaemic children after BMT were also found to have had lower diffusion capacity [27‐30]. The mechanisms leading to the persistent damage of the alveolar-capillary barrier are not clear. Recently, a model of promoting lung damage by CMV was proposed, suggesting that CMV causes a pro-fibrotic milieu by increasing the expression CCL2 that sequesters CCR2 expressing pro-fibrotic mononuclear cells of the lung [31]. The other model was also proposed that CMV may cause an increased production and activation of TGF-β1 (transforming growth factor-β1), a multifunctional cytokine playing a key role in cell proliferation, migration, and synthesis of the extracellular matrix in the endothelium [32]. TGF-β1 may contribute to the fibrotic and vascular components of CMV pathology. Interestingly, in an animal model, TGF-β1 protein was increased in alveoli in rat lungs after radiation-induced immune suppression of CMV-infected rats [33].

Our data suggest that CMV could potentially damage the alveo-capillary barrier. However, the identification of individual factors contributing to impairment of diffusion capacity in leukemic patients is difficult because of the interaction of many different interfering variables. Nevertheless, such a strong statistical correlation between decreased diffusion capacity and CMV pneumonia as it was found in our study may suggest the direct prolonged influence of CMV infection on the pulmonary tissue.

Another finding of the present study is that the HSCT procedure affected lung diffusion. Lower diffusion values were significantly more frequent in CLS who received HSCT than in the children treated with chemotherapy alone (41% vs. 14%). Similar data on DLCO dysfunction (12–28% of patients treated with chemotherapy alone and 27–58% after HSCT) was presented in other studies [22, 27, 29, 30, 33‐37]. These findings may be partially explained by the fact that the HSCT patients had a more intensive immunosuppressive treatment with additional radiotherapy. However, in our study, other factors, such as pulmonary infection, also affected these abnormalities.

A clinically meaningful finding is that a proportion of CLS (13%) had a poor tolerance of exercise, which is related to the decreased DCLO. Measurement of DLCO is considered a good indicator of pneumopathy—it can be lowered even if there were no clinical symptoms and the spirometry remains within normal values. In several studies with long follow-up, the pulmonary dysfunction (decreased diffusion capacity) was usually without clinical symptoms [27, 29, 30]. In contradiction to these studies, we observed an interesting relationship: survivors who complained of poorer tolerance of exercise had a significantly lower diffusion capacity (p < 0.005). This correlation was particularly strong in survivors, who underwent HSCT. One possible explanation is that an injury of the alveo-capillary barrier (reflected by lowered DLCO) in leukemic children treated with HSCT was more advanced, and therefore caused clinical symptoms. These symptoms, especially decreased ability to exercise, may reduce physical activity and, in turn, affect their quality of life and proper physical development.

The results of the present study show that the lower lung diffusion capacity is the most frequent abnormality in CLS. Pulmonary infection, especially caused by CMV, is a strong risk factor for impairment of lung diffusion capacity in survivors, which results in poor exercise tolerance. The particular group of risk comprised children with additional treatment including HSCT. Since the decrease of pulmonary function and the poor tolerance of exercise occur in a proportion of survivors following treatment for childhood haematologic malignancies, a screening for respiratory abnormalities in this group seems to be of importance and be part of survivorship care plan.