Mucosal barrier injury (MBI) in the gastrointestinal tract remains a major clinical obstacle in the effective treatment of hematological malignancies, driving local and systemic complications that negatively impact treatment outcomes. Here, we provide the first evidence of hyper-activation of the IL-1/CXCL1/neutrophil axis as a major driver of MBI (induced by melphalan), which supports evaluating the IL-1RA anakinra, both preclinically and clinically. Our data reinforce that strengthening the mucosal barrier with anakinra is safe and effective in controlling MBI which in turn, stabilises the host microbiota and minimises febrile events. Together, these findings represent a significant advance in prompting new therapeutic initiatives that prioritise maintenance of the gut microenvironment.
The IL-1/CXCL1/neutrophil axis is documented to drive intestinal mucosal inflammation, activated by ligation of intestinal pattern recognition receptors, including toll-like receptors (TLRs)31. In the context of MBI, TLR4 activation is known to drive intestinal toxicity32, 33, however targeting TLR4 directly is challenging due to emerging regulation of tumour response34,35,36,37. As such, we selected anakinra as our intervention to inhibit inflammatory mechanisms downstream of TLR4. While anakinra was able to minimise the intensity and duration of MBI, it did not completely prevent it with comparable citrulline dynamics across animal groups in the first 48h after melphalan treatment. This reflects the core pathobiological understanding of MBI which is initiated by direct cytotoxic events which activate a cascade of inflammatory signalling that serve to exacerbate mucosal injury and the subsequent breakdown of the mucosal barrier33. By preventing this self-perpetuating circle of injury with anakinra, we were able to effectively minimise the duration of MBI and thus have a profound impact on the clinical symptomology associated with MBI including weight loss and anorexia. These findings firstly highlight the cluster of (pre-)clinical symptoms related to MBI (malnutrition, anorexia, diarrhea)38 and suggest that the mucoprotective properties of anakinra will provide broader benefits to the host, mitigating the need for intensive supportive care interventions (e.g. parenteral nutrition).
In line with our hypothesised approach, minimising the duration of MBI reduced secondary events including enteric pathobiont expansion and fever. This again reiterates that changes in the host microbiome and associated complications can be controlled by strengthening the mucosal barrier39. It can be postulated that by minimising the intensity of mucosal injury, the hostility of the microbial environment is reduced ensuring populations of commensal microbes to be maintained. This is supported by our results with the abundance of Faecalibaculum maintained throughout the time course of MBI. Faecalibaculum is a potent butyrate-producing bacterial genus documented to control pathogen expansion by acidification of the luminal environment. Administration of Faecalibacteria prausnitzii has been shown to reduce infection load in a model of antibiotic-induced Clostridioides difficile infection, whilst also showing mucoprotective benefits in models of MBI40, 41. Furthermore, it is documented to cross feed other commensal microbes increasing colonization resistance. Together, these underscore the luminal benefits of strengthening the mucosal barrier and suggest that maintenance of commensal microbes is central to minimizing translocation events and subsequent BSI.
In our clinical Phase IIA study with 3+3 design, we have shown that treatment with anakinra, up until a dose of 300mg, appears to be safe, feasible, and tolerated well. Of course, the sample size of this study was relatively small. However, anakinra was previously evaluated for its efficacy in the treatment of acute and chronic GvHD in patients allogeneic HSCT. In these studies, patients were treated for a similar time period (with higher doses of anakinra). No differences were seen between the anakinra and placebo group regarding (S)AEs, including infections and time to neutrophil recovery. There were no significant changes in our exploratory analyses, however, it was of note to see marked increase in IL-10 in patients that received 300mg anakinra. This may reflect anakinras capacity to promote anti-inflammatory signaling as observed in COVID-19 related respiratory events42. However, with our sample size it is not possible to make any conclusions on this mechanism. Our conclusion is that the recommended dose (RP2D) for anakinra is 300mg QD, which will be investigated in Phase IIB trial (AFFECT-2 study: Anakinra: Efficacy in the Management of Fever During Neutropenia and Mucositis in ASCT; clinicaltrials.gov identifier NCT04099901)43.
While encouraging, our data must be viewed in light of some limitations. Most importantly, our animal model purposely did not include any antimicrobials as we aimed to dissect the true contribution of MBI in pathogen expansion and subsequent febrility. While it is unclear if melphalan has a direct cytotoxic effect on the microbiota, it is likely that MBI drives dysbiosis with antibiotics serving to exacerbate these changes, with previous data demonstrating no direct impact of specific chemotherapeutic agents on microbial viability44. As such, assuming dysbiosis is secondary to mucosal injury as recently demonstrated45, we anticipate that anakinra will still have an appreciable impact on the severity of dysbiosis and may even prompt more protocolised/limited antibiotic use. Similarly, while we used body temperature as an indicator of BSI, we did not culture peripheral blood or mesenteric lymph nodes as was performed in our animal model development. The ability of anakinra to prevent BSI and thus minimise antibiotic use will be best evaluated in AFFECT-2 where routine blood culture is performed. It is also important to consider that we detected episodes of bacteremia in our participants that were likely caused by skin colonizing organisms; a mechanism anakinra will not influence. While these are expected in HSCT recipients, the majority of infectious cases originate from the gut, and we therefore anticipate anakinras capacity to strengthen the mucosal barrier will be clinically impactful in our next study. It must also be acknowledged that limited mechanistic investigations were conducted to identify the way in which anakinra provided mucoprotection. It is well documented that MBI is highly multifactorial, involving mucosal, microbial and metabolic dysfunction33, 46; each of which is mediated through aberrant cytokine production. It is therefore unlikely that anakinra will affect distinct pathways, instead dampening multiple mechanisms. In translating this evidence to the clinic, the impact of anakinra on symptom control is of greater significance than mechanistic insight.
In conclusion, we have demonstrated that not only is anakinra safe in HSCT recipients treated with HDM, but may also be an effective strategy to prevent acute MBI. Our data are critical in supporting new antibiotic stewardship efforts directed at mitigating the emerging consequences of antibiotic use. We suggest that minimizing the severity and duration of MBI is an important aspect of infection control that may optimize the efficacy of anti-cancer treatment, decreasing its impact on antibiotic resistance and the long-term complications associated with microbial disruption.
This study is reported using the ARRIVE guidelines for the accurate and reproducible reporting of animal research.
All animal studies were approved by the Dutch Centrale Commissie Dierproeven (CCD) and the Institutional Animal Care and Use Committee of the University Medical Centre Groningen, University of Groningen (RUG), under the license number 171325-01(-002). The procedures were carried out in accordance with the Dutch Experiments on Animals (Wet op de Dierproeven) and the EU Directive 2010/63/EU. All animals were individually housed in conventional, open cages at the Centrale Dienst Proefdieren (CDP; Central Animal Facility) at the University Medical Centre Groningen. Rats (single housed) were housed under 12h light/dark cycles with ad libitum access to autoclaved AIN93G rodent chow and sterile water. All rats acclimatised for 10days and randomised to their treatment groups via a random number sequence generated in Excel. Small adjustments were made to ensure comparable body weight at the time of treatment and cages were equally distributed across racks to minimise confounding factors. HRW was responsible for animal allocation and assessments while RH/ARDSF performed treatments. Softened chow and subcutaneous saline were provided to rats to reduce suffering/distress and were humanely euthanised if a clinical toxicity score>/=12 was observed. This score was calculated based on weight loss, diarrhea, reluctance to move, coat condition and food intake; each of which were assessed 03. At completion of the study, rats were anaesthetised with 5% isoflurane in an induction chamber, followed by cardiac puncture and cervical dislocation (isoflurane provided by a facemask).
We have previously reported on the development and validation of our HDM model of MBI, which exhibits both clinical and molecular consistency with patients undergoing HDM treatment21. During model development, plasma (isolated from whole blood) was collected and stored for cytokine analysis to inform the selection of our intervention. Repeated whole blood samples (75l) were collected from the tail vein into EDTA-treated haematocrit capillary tubes on day 0, 4, 7 and 10.
Cytokines (IFN-, IL-1, IL-4, IL-5, IL-6, IL-10, IL-13, KC/GRO and TNF-) using the Meso Scale Discovery V-Plex Proinflammatory Panel Rat 2 following manufacturers guidelines. On the day of analysis, all reagents were brought to room temperature, samples were centrifuged to remove any particulate matter and diluted 1:4. Data analysis was performed using the Meso Scale Discovery Workbench.
Male albino Wistar rats (150180g) were randomized (Excel number generator) to one of four experimental groups (N=16/group): (1) controls (phosphate buffered saline (PBS)+0.9% NaCl), (2) anakinra+0.9% NaCl, (3) PBS+melphalan, and (4) anakinra+melphalan. Melphalan was administered as a single, intravenous dose on day 0 (5mg/kg, 10mg/ml) via the penile vein under 3% isoflurane anaesthetic. Anakinra was administered subcutaneously (100mg/kg, 150mg/ml) twice daily from day 1 to+4 (8 am and 5pm). N=4 rats per group were terminated at the exploratory time points (day 4, and 7) and N=8 on day 10 (recovery phase) by isoflurane inhalation (3%) and cervical dislocation. The primary endpoint for the intervention study was plasma citrulline, a validated biomarker of MBI19, 47, which was used for all power calculations (N=8 required, alpha=0.05, beta=0.8).
Clinical manifestations of MBI were assessed using validated parameters of body weight, food intake and water intake, as well as routine welfare indicators (movement, posture, coat condition). Rats were weighed daily, and water/food intake monitored by manual weighing of chow and water bottles.
Plasma citrulline is an indicator of intestinal enterocyte mass48, and a validated biomarker of intestinal MBI. Repeated blood samples (75l) were collected from the tail vein into EDTA-treated haematocrit capillary tubes on day 0, 2, 4, 6, 7, 8 and 10. Citrulline was determined in 30l of plasma (isolated from whole blood via centrifugation at 4000g for 10min) using automated ion exchange column chromatography as previously described49.
Whole blood samples (200l) were collected from the tail vein into MiniCollect EDTA tubes on day 0, 4, 7 and 10 for differential morphological analysis which included: white blood cell count (WBC, 109/L), red blood cell count (RBC, 109/L), haemoglobin (HGB, mmol/L), haematocrit (HCT, L/L), mean corpuscular volume (MCV, fL), mean corpuscular haemoglobin (MCH, amol), mean corpuscular hemoglobin concentration (MCHC, mmol/L), platelet count (PLT, 109/L), red blood cell distribution width (RDW-SD/-CV, fL/%), mean platelet volume (fL), mean platelet volume (MPV, fL), platelet large cell ratio (P-LCR, %), procalcitonin (PCT, %), nucleated red blood cell (NRBC, 109/L and %), neutrophils (109/L and %), lymphocytes (109/L and %), monocytes (109/L and %), eosinophils (109/L and %), basophils (109/L and %) and immunoglobulins (IG, 109/L and %). For the purpose of the current study only neutrophils, lymphocytes and monocytes were evaluated.
Core body temperature was used as an indicator of fever. Body temperature was assessed daily using the Plexx B.V. DAS-7007R handheld reader and IPT programmable transponders. Transponders were inserted subcutaneously under mild 2% isoflurane anaesthesia on day 4. Average values from day 4 to 1 were considered as baseline body temperature.
The microbiota composition was assessed using 16S rRNA sequencing in N=8 rats/group. Repeated faecal samples were collected on day 0, 4, 7 and 10 and stored at 80C until analysis. Sample preparation (including DNA extraction, PCR amplification, library preparation), quality control, sequencing and analyses were all performed by Novogene (please see supplementary methods for full description).
All data (excluding 16S data) were analysed in GraphPad Prism (v8.0. Repeated measures across multiple groups were assessed by mixed-effect models with appropriate post-hoc analyses. Terminal data analyses were assessed by one-way ANOVA. Statistical analyses are outlined in figure legends and P<0.05 was considered significant.
This Phase IIA trial (AFFECT-1: NCT03233776, 17/6/2017) aimed to i) assess the safety of anakinra in autologous HSCT recipients undergoing conditioning with HDM, and ii) determine the maximum tolerated dose of anakina (100, 200 or 300mg).
This study was approved by the ethical committee Nijmegen-Arnhem (NL59679.091.16; EudraCT 2016-004,419-11) and performed in accordance with (a) theDeclaration of Helsinki (1964, amended October 2013), (b) Medical Research Involving Human Subjects Act and c) Good Clinical Practice guidelines.We enrolled patients from Radboud University Medical Centre who were at least 18years of age and were scheduled to undergo an autologous HSCT after receiving conditioning with HDM (200mg/m2) for multiple myeloma. All participants provided informed consent. Important exclusion criteria were active infections, a history of tuberculosis or positive Quantiferon, glomular filtration rate<40ml/min, and colonization with highly resistant micro-organisms or with gram-negative bacteria resistant to ciprofloxacin.
Patients were involved in the design of the AFFECT trials, through involvement of Hematon, a patient organization for patients with hemato-oncological diseases in the Netherlands. The project plan, including trial materials, have been presented to patient experts from Hematon. They have given their advice on the project, and provided input on the design of the study as well as on patient information. Patients will also be involved in the dissemination of the results of the AFFECT trials. Information on both the design as well as the outcome of the AFFECT trials is and/or will be available on websites specifically aimed at patients, such as the Dutch website kanker.nl.
Conforming with routine clinical practice and care, study participants were admitted at day 3, treated with melphalan 200mg/m2 at day 2, and received their autologous HSCT at day 0. They were treated with IL-1RA anakinra (Kineret, SOBI) intravenously once daily from day 2 up until day+12.
A traditional 3+3 design was used (Fig. S1), in which the first cohort of patients was treated with 100mg, the next cohort with 200mg and the third cohort with 300mg of anakinra. In this study design, the cohort is expanded when dose limiting toxicities (DLTs) occur. The primary study endpoint was safety, using the common toxicity criteria (CTCAE) version 4.050, as well as the maximum tolerated dose of anakinra (MTD; 100, 200 or 300mg). DLTs were defined as the occurrence of (1) an infection due to an opportunistic pathogen (including Pneumocystis jirovecii pneumonia, mycobacterial infections and invasive mould disease), (2) a suspected unexpected serious adverse reaction (SUSAR), (3) severe non-hematological toxicity grade 34 (meaning toxicity that does not commonly occur in the treatment with HDM and HSCT, or that is more severe than is to be expected with standard treatment) and (4) primary graft failure or prolonged neutropenia (neutrophils have not been>0.5109/l on one single day, assessed on day+21, and counting from day 0).
Secondary endpoints included: incidence of fever during neutropenia (defined as a tympanic temperature38.5C and an absolute neutrophil count (ANC)<0.5109/l, or expected to fall below 0.5109/l in the next 48h), CRP levels, intestinal mucositis as measured by (the AUC of) citrulline, clinical mucositis as determined by daily mouth and gut scores, incidence and type of BSI, short term overall survival (100days and 1year after HSCT), length of hospital stay in days and use of systemic antimicrobial agents, analgesic drugs and total parenteral nutrition (incidence and duration).
Patients received standard antimicrobial prophylaxis including ciprofloxacin and valacyclovir, as well as antifungal prophylaxis (fluconazole) on indication; i.e. established mucosal colonization. Upon occurrence of fever during neutropenia, empirical treatment with ceftazidime was started. The use of therapies to prevent or treat mucositis (i.e. oral cryotherapy) was prohibited. Also, treatment with acetaminophen or non-steroidal anti-inflammatory drugs was not allowed during hospital admission. All other supportive care treatments (i.e. morphine, antiemetics, transfusions, TPN) were allowed.
Laboratory analysis was performed three times a week, which included hematological and chemistry panels and plasma collection for citrulline analysis. Blood cultures were drawn daily from day+4 up until day+12, which was halted upon occurrence of fever. Outside this period, conforming to standard of care, blood cultures were drawn twice weekly and in occurrence of fever. Conforming standard of care, surveillance cultures of mucosal barriers were obtained twice weekly.
Plasma was longitudinally collected from participants throughout the study period for the evaluation of cytokines using the Meso Scale Discovery Customised U-Plex 9-analyte panel following manufacturers guidelines (IL-1/, IL-1RA, CXCL1, TNF, IL-10, IL-17, IL-6, GM-CSF). 16S sequencing was performed by Novogene (as per preclinical analysis methodology).
Continue reading here:
Supporting the gastrointestinal microenvironment during high-dose chemotherapy and stem cell transplantation by inhibiting IL-1 signaling with...
