- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
Leukemia is the most common pediatric cancer and leading cause of cancer related deaths in children. Improvements in the assessment of leukemic cells have the potential to influence not only the diagnosis of leukemia, but also the risk assessment of patients during the course of the treatment, both of which are important for improving the cure rate for this disease. In this study, we report on the design and performance of a confocal laser based system built to collect backscattered light over a range of 26° at 405, 488, and 633 nm to discriminate leukemic cells from normal red blood cells (RBC) and white blood cells (WBC). The design of the system is based on the spectral differences observed from spectroscopy measurements with a similar system designed with a white light source. Significant differences are observed in the intensity and wavelength dependence of leukemic cells from normal RBC and WBC. Specifically, the distinct light scattering of RBC is due to hemoglobin absorption, allowing for its discrimination from leukemic cells, mononuclear, and polymorphonuclear WBC particularly at certain wavelengths. Meanwhile, the high scattering intensities of polymorphonuclear WBC reflect the intracellular complexity of these cells in comparison to the leukemic or normal lymphocytes. Additionally, the detected light scattering spectra for leukemic cells are consistently steeper in comparison to normal WBC, which we attributed to differences in the fractal organization of intracellular scatterers. Based on our findings, the system has potential applications in the detection and quantification of leukemic cells in blood either in vivo or in vitro, using microfluidic-based systems, for disease monitoring. © 2011 International Society for Advancement of Cytometry
Hematopoiesis, which describes the production of mature blood cells from stem cells in the bone marrow, is a highly regulated and precise process. Disruption in normal hematopoiesis can result in serious conditions such as leukemia, the most common blood cancer. Leukemia is characterized by the overproduction of malignant and immature leukocytes that are unable to carry out normal hematopoietic functions, resulting in clinical manifestations related to infectious or hemorrhagic complications (1). It is expected to affect more than 44,000 adults and ∼3,500 children under the age of 15 in the United States (2). Although leukemia affects more adults than children, it is the most common pediatric cancer, accounting for 25% of cancer cases, and is the leading cause of disease related death in children under 20 years of age (3, 4). Acute lymphoblastic leukemia (ALL), the most common form of leukemia in children, involves the increased proliferation of lymphocytes arrested in the early stage of development, resulting in overcrowding in the bone marrow and their eventual presence in the circulation, where they are not commonly found.
Blood analysis has been found to be of clinical importance in the prognostic assessment of patients after therapy, with patients showing a persistent number of leukemic blasts in the bone marrow or peripheral blood associated with slow therapeutic response and a high risk for relapse (5–7). Morphological analysis of cell samples, which is the traditional technique for assessing the number of residual leukemic cells that represent minimal residual disease (MRD), is subjective and limited in sensitivity (8). Thus, a considerable proportion of patients who are considered to be in remission, defined by a morphological criterion of having less than 5% blast cells in the bone marrow, may in fact harbor as many as 1010 residual leukemic cells and eventually experience a relapse (8, 9). To improve the assessment of MRD for monitoring therapeutic response, methods, such as flow cytometric detection of aberrant immunophenotypes and polymerase chain reaction (PCR) analysis for gene rearrangement, have been developed (10–12). Nonetheless, the applicability of flow cytometry or PCR is limited and depends on the type of ALL and MRD assay used (9, 13–16). Moreover, false positives may result not only from immunophenotypic changes or clonal gene rearrangements, but also from practical differences in laboratory protocols, including sample handling, flow cytometry gating or choice of PCR reference genes (10, 14, 17). Therefore, improvements in the detection of leukemic cells in blood or bone marrow aspirates can have a significant clinical impact, by improving MRD assessment and allowing effective therapeutic intervention.
Optical techniques have the potential to yield improved or novel approaches to detect leukemia and monitor MRD. For example, in vivo flow cytometry, which combines the principles of confocal detection and flow cytometry, has been used for the detection of fluorescently labeled cancer cells directly in the circulation of animals and has provided insights in terms of the kinetics of cancer cells and metastasis (18–25). However, since few dyes are approved for use in humans, techniques that are based on natural sources of signal contrast may have more clinical applicability. Specifically, intrinsic light scattering is sensitive to cellular and subcellular morphology and organization and has been used for the discrimination of different cell types. The angular dependence of light scattering has been shown to differentiate between sub-populations of white blood cells (WBC) in a flow cytometry set-up (26–29). In the context of cancer, light scattering spectroscopy has identified differences in the wavelength dependence of backscattered light between normal and diseased cells due to cellular transformations in early stages of cancer development (30–36). Despite these studies, little work has explored the potential of light scattering for the identification of leukemic cells from blood cells for leukemia diagnosis or MRD assessment. With this in mind, we developed light scattering based systems to characterize the backscattering of leukemic and normal white blood cells (WBC) and red blood cells (RBC). These systems provide the basis for the design of in vitro microfluidic-based and potentially in vivo light scattering flow cytometry devices.
- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
In our study, we characterized the backscattering of leukemic cells, WBCS, and RBC at various wavelengths. This approach takes advantage of the light scattering spectral differences of these populations that are attributed to hemoglobin (for the differentiation of RBC from non-RBC) as well as differences in the morphology and organization of the cells, especially in the context of separating leukemic from normal WBCs.
Hemoglobin absorption effects were observed in both ABSS and LBSS measurements of RBC. Notably, hemoglobin absorption resulted in the lowest detected intensities for RBC for wavelengths up to 600 nm. At longer wavelengths, higher scattering was observed for RBC than for normal PBMC and cancerous Nalm-6 cells. These results suggest that the choice of wavelength can be critical in cell discrimination. Indeed, this was noted in a flow cytometry study that showed improved discrimination of WBC sub-populations from RBC at a shorter wavelength, namely 413 nm, compared with 488 nm (29). By selecting wavelengths associated with low, medium, and high hemoglobin absorption, as in our LBSS design, RBC may readily be identified from both normal and cancerous WBCs.
Meanwhile, morphological scale invariance is evident by the inverse power-law spectral dependence of scattering, likely reflecting some type of fractal subcellular morphology especially in diseased WBCs. Although the exact form of the fractal nature (e.g., self-affine vs. mass fractal geometry) cannot be uniquely inferred by our current light scattering experiments, we note that changes in the light scattering power exponent, γ, have been correlated to scale invariant subcellular inhomogeneities observed by differential interference contrast (DIC) microscopy on these cells (42). Similar power-law spectral dependence has been observed for leukemic and normal WBCs in a previous study by our group that characterized the scattering at narrower angles, (0° ± 4° from the exact backscattering angle), but with γ values that are higher in comparison to our study (42). Specifically, the wavelength dependent exponent values for fits to spectra formed by light scattered at 1° relative to exact backscattering angle from the previous study are γ = 1.83 ± 0.07, γ = 0.86 ± 0.10, and γ = 0.7 ± 0.04 for Nalm-6, PBMC, and PMN, respectively. Since our spectra represent the scattered light over a large range of angles (0–26°) as opposed to a specific angle (i.e., 1°), the differences in the exponents suggest that light scattering at higher angles exhibits a more shallow wavelength dependence, resulting in the overall flatter wavelength profiles observed in our study. Nevertheless, we should note that in both studies, the leukemic cells exhibit significantly steeper wavelength dependence than the two normal leukocyte populations, which in turn are characterized by similar wavelength decays. Therefore, our results show significant promise for the light scattering for the non-invasive or minimally invasive diagnosis of leukemic conditions, as well as for providing insight into the subcellular morphological transformations associated that give rise to the observed differences.
In addition to spectral differences, consistent and significant differences are observed in the overall intensity of the detected light in both ABSS and LBSS measurements. Specifically, the scattering intensity from PMN is highest at all wavelengths and lowest for PBMC compared with leukemic cells. The high total scattering intensity from PMN maybe due to the large number of granules found in the cytoplasm of these cells, which are not present in PBMC or Nalm-6 cells (38, 43–45). These differences in the scattering intensity of spectra from normal and leukemic WBCs are also consistent with our previous study using a collimated illumination system (42).
In summary, our main findings indicate that leukemic cell discrimination from cells most commonly found in blood is possible with light scattering spectroscopy, based on a system design employing focused illumination, confocal detection, and collection of light over a broad range of backscattering angles, at a small number of excitation wavelengths spanning the visible range. Therefore, light scattering based approaches may result in the development of novel, minimally or noninvasive techniques, such as microfluidic-based lab-on-chip or in vivo flow cytometry platforms with confocal detection, to improve monitoring of leukemic cells for disease diagnosis or MRD assessment particularly in children, for whom blood withdrawal maybe difficult and traumatic.