In cancer patients, myelotoxicity is the most frequent side-effect of cytostatic therapy [1]; it puts the patient at risk for developing septic or bleeding complications and often results in delays in planned chemotherapy [2]. For this reason, understanding the mechanism of bone marrow toxicity is an important step toward improving supportive care in cancer treatment. Measurement of platelet RNA offers the opportunity to quantify the process of bone marrow regeneration after chemotherapy [11–14]. In fact, automatic hematology analyzers detect immature platelets easily and their quantification correlates with the rate of thrombopoiesis [10–13, 16–18]. At first glance, it is not obvious how thrombopoiesis plays a role in predicting bone marrow recovery, but IPF% has already been recognized as a parameter of imminent thrombocyte regeneration following cytostatic treatment [19]. Whether the data on IPF% can predict the rise in other peripheral blood cells after chemotherapy is still unknown.
The example given in Fig. 1A-I shows the typical course of peripheral blood cell counts in a patient after initial cytoreductive treatment. The optimal management in this situation requires a detailed anamnesis and physical examination – with particular attention given to the skin, catheter sites, lungs, sinuses, mouth and abdomen. In the case of a patient with unclear fever and neutropenia, immediate empiric therapy with a dose of intravenous antibiotics is required. Once the patient is improving and blood cultures remain negative, treatment can be de-escalated to more narrow coverage. Whether the patient is continued on this therapy before neutropenia has resolved, is controversial and practices vary among centers if no infectious etiology is identified. Conversely, the responsible clinician will closely evaluate the patient for signs of neutrophil recovery. In this example we demonstrate that an increase in IPF% anticipates regeneration of neutrophilic granulocytes and suggests that discontinuation of antibiotics before neutrophil recovery in a stable patient with recognition of increasing IPF% may be a reasonable approach.
Although we anticipate that more data on this topic are necessary, a much broader investigation of the normal levels of IPF% is needed. Numerous studies limit the analysis of IPF% to healthy and adult patients and show values between 0.3% and 17.8% [20]. The most accurate studies are performed in accordance with the CLSI (Clinical and Laboratory Standards Institute) guidelines in very large study populations and result in IPF% values between 0.5%-3.2% and 0.4%-3% [21]. In pediatrics, large studies in different age groups are missing. One single study limits the analysis to the neonatal period and attributes distinct values to different gestational ages, with the values differing between 1.5% and 5.9% [20]. Another study limits the IPF% values (0.7%-5.7%) to 100 children between 6 months and 18 years of age and includes samples from patients with a wide variety of different diagnoses [9]. Our analysis of IPF% in 416 healthy children shows that IPF% is sex-independent and differs concerning the age (Table 1). Infants have the highest IPF% in comparison to other age groups. Although the comparison is made with a small number of patients, this correlates with previous data showing higher IPF% in children younger than 4 years of age [9]. It is known that thrombopoietin has higher concentration levels in this age group, which translates to increased activity of megakaryocytes and production of thrombocytes [9, 24, 25].
Considering that IPF% is a useful diagnostic parameter for identifying the production of thrombocytes [8, 13], it has long been claimed to have its value in predicting neutrophil regeneration after chemotherapy [26]. After observing the course of cell count in an individual case, we analyzed 11 uniformly treated patients diagnosed with Ewing sarcoma. The nadir of neutrophil, reticulocyte and platelet count resulted early after initiation of chemotherapy and was immediately paralleled by a simultaneous rise in IPF%. Interestingly, in our study the IPF% peak occurred almost simultaneously with the CRP peak (Fig. 2a). In the observed cases, the increase in peripheral blood cells followed a little later, in the order of platelets, neutrophilic granulocytes and reticulocytes. This sequence makes sense for the expected regeneration of bone marrow after chemotherapy, as platelets have the shortest proliferation time and reticulocytes have the longest time. On the other hand, caution is advised for this interpretation, as platelets are destroyed and IPF% is subsequently increased in patients with sepsis, regardless of the treatment with chemotherapy [18, 27, 35]. In this sense IPF% has been proposed as a biomarker for the prediction of sepsis diagnosis and severity [37][39, 40].
Nonetheless, the findings presented and discussed here must be interpreted with caution and a number of limitations should be borne in mind. Most of the studies, including ours, were performed retrospectively and possibly include a biased selection of controls. Our and other studies may underestimate the IPF% values, as this parameter appears very quickly and disappears at the same time. Furthermore, IPF% measurement is not standardized and is performed with different devices. The only way to overcome these limitations is to perform a prospective clinical study with a large number of subjects, well-defined methods and precisely formulated aims.
We conclude that analysis of IPF% may give helpful data on neutrophil activity and kinetics in the post-chemotherapy period in children. Considering these data may help determine timing and the need for antibiotic therapy in phases of aplasia. We suggest that this method be included in routine clinical tools in pediatric oncology, as it is not expensive, readily available and a precise method for estimating bone marrow kinetics.