Roy Spijkerman

Fully automated flow cytometry in a field laboratory is used to analyze patient with inflammatory bowel disease (ulcerative colitis and Crohn's disease). This study shows that decreased responsiveness of neutrophils and monocytes to fMLF was demonstrated after repetitive bouts of prolonged exercise in these patients (chapter 6). These refractory cells might create a lower inflammatory state in the intestine, providing a putative mechanism for the decrease in flare-ups in these patients after repeated exercise.

Fully automated flow cytometry also enabled clinical applicability in daily patient care. Chapters 7, 8 and 9 not only investigated the ‘near-real’ in-vivo cell biology but also provided clinical correlations with specific cell markers.

Point-of-care fully automated flow cytometry was used to study neutrophil responses in trauma patients in chapters 8 and 9. Implementation of point-of-care fully automated flow cytometry in the trauma room appeared feasible (chapter 9). Neutrophil phagosomal acidification differs between patients who develop infectious complications and patients who do not (chapter 8). The assessment of CD16dim/CD62Lbright neutrophils is used for early detection of patients at risk for infectious complications (AUC = 0.90) (chapter 9). The %CD16dim/CD62Lbright neutrophils provided valuable information for clinical decision making in trauma patients (chapter 9). With the results of chapter 9, the trauma surgery department of the UMC Utrecht has decided to implement this analysis as a standard-of-care procedure for support in clinical decision making. To further elaborate the results of chapter 9 in a multivariate model, an international multicenter study is initiated. Chapter 10 shows a study protocol for the development and testing of a multivariate prediction model in a multicenter study.

General discussion, future perspectives and conclusions

General discussion

‘Near-real’ in-vivo status of the human innate immune system

Investigating the ‘near-real’ in-vivo status of innate immune cells has always been challenging in the study of the peripheral blood of humans. It seems essential to measure cells in a situation as close to the in-vivo situation as possible, to understand cellular behavior in situ better. However, innate immune cells are very susceptible to both (bio)chemical and physical stimuli. In the ideal case, innate immune cells need to be measured without ex-vivo manipulation. At this moment, in-vivo fluorescent labeling with in-vivo imaging techniques would be the closest readout of the ‘real’ in-vivo status of innate immune cells. Several animal studies showed 1,2 exciting results using this method, albeit that the clinical procedures needed for this technique (e.g. insertion of a window) will lead to a breach in homeostasis 3,4 leading to responses of innate immune cells. In-vivo imaging of human innate immune cells is only performed in a few studies, because of its complexity 5.

The next best thing, that provides available and accessible readout of the innate immune system is analyzing minimally manipulated human peripheral blood. In the early days of cellular analysis researchers relied on long procedures to isolate cells 6,7 as pure as possible. Later it became clear that a massive number of phagocytes die and/or activate within a few hours during these general work-up procedures. Therefore, multiple specific work-up protocol were designed for the adequate analysis of phagocytes 8. However, this new work-up protocol did not have the availability of modern technology. We demonstrated that when analyzing phagocytes, the in-vitro work-up as generally used today, still induces an enormous in-vitro bias, compared to automated flow cytometry without in-vitro work-up (chapter 3).

From the moment that the needle enters the skin during blood drawing, innate immune cells will be activated. Multiple DAMPs will likely be released by the skin, subcutis and vein wall, which results in quick activation of innate immune cells expressing specific receptors for these DAMPs 9,10. Furthermore, the different shear stress of the blood flowing through the needle changes the physiological circumstances for the cells. In addition, the blood is collected in a blood tube with anti-coagulants, again changing the homeostatic conditions of the cell environment. Unfortunately, the steps mentioned above are unavoidable when collecting human blood. Then the ‘time starts ticking’ (chapter 3), where every minute waiting means more cell stress and more cell activation.

All studies regarding neutrophils activation determined by flow cytometry executed before this thesis used a conventional manual flow cytometry method to investigate the activation of innate immune cells ex-vivo 11–15. With these manual methods, cells will undergo significant but unavoidable physiological stress due to multiple manipulation steps: transferring cells to different tubes, manual pipetting, spinning in centrifuge, vortexing and washing. The time that all these steps take, vary from sample to sample, but mostly take a minimum of 2 hours. With the knowledge from chapter 3 (significant artificial activation within 30 min), one can imagine that the neutrophil activation status is then far from the ‘near-real’ in-vivo activation status of the innate immune cells. Unfortunately, before introducing fully automated flow cytometry, it was hardly possible to set up a study as chapter 3. With this knowledge, we were able to come one step closer to investigate the ‘near-real’ in-vivo activation status.

Moreover, it is not only crucial for investigating ‘near-real’ in-vivo cell biology but is also very important for clinical studies. Measuring neutrophil activation markers of healthy controls with a throughput time of 2 to 3 hours, compared with a disease group with a (mostly longer) throughput time of 4 to 5 hours, will always result in a significant increase in neutrophil activation markers in the samples to take the longest time to analyze. Having this knowledge, many studies (not reporting lab throughput time) published before should be interpreted with great caution 13,15.

Fast laboratory logistics versus point-of-care system

All conventional flow cytometers and work-up/analysis protocols require specialized flow cytometry laboratories and specific knowledge. Fully automated flow cytometry has the huge advantage that there is no dedicated laboratory nor specialized personnel necessary to measure blood samples. Fully automated flow cytometry enables the point-of-care implementation of this flow method. This fits perfectly as a hands-on clinical decision support system in daily hospital practice. We show in chapter 9 that it is feasible to implement this method in the trauma room, with only nurses and medical doctors handling the samples. In the future, it will also be possible to place the machine in another hospital ward, intensive care unit, emergency department or outpatient clinic. The most important is to ensure minimal ex-vivo time bias.

Point-of-care fully automated flow cytometry has the advantage of really bringing bench-to-bedside and can provide an easy accessible clinical decision support system. However, point-of-care positioning has also downsides. The machine needs to have a physical place, troubleshooting needs to be done, the machine makes some noise and maybe the most important issue is that only one department can make use of the machine. Placing the machine in a central diagnostic laboratory with 24/7 availability and fast laboratory logistics would circumvent the above-mentioned problem. However, the central diagnostic laboratory needs to ensure optimal logistics and a very fast handling of blood samples. This is not the case in many diagnostic labs in the world. With all new research and biomarkers in the scientific area of flow cytometry, the central diagnostic labs (especially in an academic hospital) need to keep up with new developments.

Clinical decision making in trauma surgery

The standard work-up of trauma patients in the western world is physical examination, radiography, ECG and blood testing to get an overview of the clinical characteristics caused by the injuries 16. If there is enough time, a CT-scan will also be made. These diagnostic techniques will help the trauma surgeon in clinical decision making, with patient survival as primary priority.

More and more patients survive the initial trauma nowadays due to advances in the prehospital trauma system, (surgical) hemorrhage control and resuscitation 17,18. Therefore, the main reason for in house mortality of trauma in the western world has changed towards mortality due to immune-related complications, such as an overwhelming immune responses, severe infections, or recurrent infections later on 19,20. Therefore, the priorities in clinical decision-making nowadays shifted towards preventing immune-related complications. The first step in this process is recognizing patients at risk for developing immune-related complications.

Chapter 9 provides a biomarker, %CD16dim neutrophils, that correlated with the Injury Severity Score that is associated with the cumulative amount of tissue damage in trauma patients. This biomarker also had significant value in the prediction of infectious complications in polytrauma patients. With the support of the %CD16dim neutrophils, it becomes easier to predict which patients are at risk for developing immune-related complications. With this knowledge, the clinician can decide whether treatment needs to be according to an immune-protective protocol or according to an early recovery protocol. See figure below.

Figure 1: Percentage of CD16dim/CD62Lbright neutrophils against the ISS in all included trauma patients. Data are presented as a scatter plot with the linear regression line and 95% confidence interval (R2=0.43, P<0.0001). All patients with an injury severity score <16 had <6% CD16dim neutrophils, except for one outlier with 12% CD16dim neutrophils (left red square). This trauma patient was presented after an out-of-hospital cardiac arrest. All polytrauma (ISS≥16) patients had CD16dim neutrophils varying from 1% to 22%. The second outlier is a polytrauma patient with an ISS of 75, with only 6% CD16dim/CD62Lbright neutrophils (right red square). This was a patient with a very high-grade head and a non-survivable neck injury with little additional trauma. Patients with a >6% CD16dim neutrophils are at risk for immune-related complications and should be treated according to the immune-protective protocol. Patients with <6% CD16dim neutrophils can be treated with the early recovery protocol. The immune protective protocol entails all supportive care to prevent immune-related complications. Supportive care in these trauma patients at risk for immune-related complication can be initial damage control surgery, adequate timing of definitive surgery, prophylactic antibiotics and enhanced nutritional support. If patients are not at risk for immune-related complications, the main goal of treatment is early recovery and soon discharge. As the next step based on chapter 9, the UMC Utrecht has implemented fully automated flow cytometry as a standard-of-care diagnostic modality. In every trauma patients who presents in the trauma room, neutrophil phenotype is measured and the %CD16dim neutrophils is automatically sent to the electronic patient registry. The results of this test, in combination with all other clinical parameters, helps in better clinical decision-making and improved patient care. Future perspectives Flow cytometry has been used to analyze cells and particles for a long time now and has resulted in exciting and intriguing results 13,15,21–23. However, the next step, bringing biomarkers determined by flow cytometry to clinical diagnostics, was always a difficult step (chapter 2). With point-of-care fully automated flow cytometry, it is now possible to more easily bring bench to bedside. This will be illustrated by the following examples: Personalized surgery using fully automated flow cytometry Immune-related complications belong to the most common clinical issues seen after surgery in the western world. A balanced innate immune system is essential for minimizing immune-related complications (chapter 2). For every patient, this balance is different and should, therefore, be individually assessed. With fully automated flow cytometry, it is easier to implement flow cytometry determined markers to daily clinical practice and for personalized medicine. In chapter 9, we have identified a valuable marker for the prediction of infection in polytrauma patients. A few months after publication of this study, the trauma surgery department started using this biomarker as standard-of-care practice. This is only one example of how easy it can be to bring bench to bedside using fully automated flow cytometry. There are many possible applications where innate immune analysis by fully automated flow cytometry can improve personalized surgery: 1. Preoperative screening of patients undergoing surgery to check innate immune activation status. (e.g. screening before surgery) 2. Determine the effect of surgery on a patient’s immune system and reduce infectious complications (e.g. determination of maximum surgery time per patient, immune-monitoring during surgery) 3. Find the optimal timing for surgery to minimize infections (e.g. timing of definitive fixation after trauma) 4. Determine operation indication based on the immune system (e.g. treatment of hip fracture in elderly patients) 5. Differentiate between viral infection, bacterial infection or sterile inflammation (e.g. daily determination of CD64) 6. Immune monitoring on the ICU (e.g. determination of neutrophil acidification) The ‘immuno’ bus: bringing the lab to the people There are also many more possibilities for implementing fully automated flow cytometry outside the hospital in research and clinical setting. The machine is ‘dummy-proof’ and stable and, therefore, the device could be positioned in a field laboratory or bus. Chapters 6 and 7 showed examples of studies done in a field laboratory at a sport event. Implementing the machine in an ‘immuno’ bus, will result in even more research and clinical possibilities, without losing time in samples analysis when analyzing innate immune cells (chapter 3). A few examples of possible implementations for the ‘immuno’ bus: 1. Position flow analysis at sports events to investigate the innate immune system in a large group of athletes. 2. Support clinical trials in providing home visits during pandemics (e.g. COVID-19) 3. Creating a point-of-care HIV-lab traveling to provide care in rural areas with large distances to a hospital in Africa or South-America. Next-generation fully automated flow cytometers In this thesis, we used the AQUIOS CL fully automated flow cytometer. There are some possible competitors brought to the market last few years providing flow cytometers with prep stations (e.g. BD FACSDuet™, MACSQuant®, Laminar Wash™ AUTO 1000), however none of them were so dummy-proof and “all-in one” as the AQUIOS CL. However, the AQUIOS CL has several technical limitations: 1. It has only one laser, 2. five ‘relatively old’ sensors, 3. is relatively large and 4. is a closed system (which is also an advantage as it is dummy proof). The ideal next-generation fully automated flow cytometers need to have multiple lasers, several high-resolution sensors, are small and adjustable by the programmer. A combination of the technical specs of the CytoFLEX (Beckman coulter) in combination with the AQUIOS CL will be a significant step forwards in analysis of innate immune cells in the future. Conclusions Fully automated flow cytometry is a new method for the analysis of innate immune cells. This new method improves the measurement of innate immune cells and leads to a better understanding of the ‘near-real’ in-vivo status of these cells. Moreover, this new technique enables point-of-care implementation and relatively easy clinical applicability. New biomarkers assessed with this machine can quickly be brought from bench to bedside. Implementation in daily clinical practice will allow fully automated flow cytometry-based precision medicine.