Defining and Grading Infections in Clinical Trials Involving Hematopoietic Cell Transplantation: A Report from the BMT CTN Infectious Disease Technical Committee

Abstract

The Blood and Marrow Transplant Clinical Trials Network (BMT-CTN) was established in 2001 to conduct large multi-institutional clinical trials addressing important issues towards improving the outcomes of HCT and other cellular therapies. Trials conducted by the network investigating new advances in HCT and cellular therapy not only assess efficacy but require careful capturing and severity assessment of adverse events and toxicities. Adverse infectious events in cancer clinical trials are typically graded according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events (CTCAE). However, there are limitations to this framework as it relates to HCT given the associated immunodeficiency and delayed immune reconstitution. The BMT-CTN Infection Grading System is a monitoring tool developed by the BMT CTN to capture and monitor infectious complications and differs from the CTCAE by its classification of infections based on their potential impact on morbidity and mortality for HCT recipients. Here we offer a report from the BMT CTN Infectious Disease Technical Committee regarding the rationale, development, and revising of BMT-CTN Infection Grading System and future directions as it applies to future clinical trials involving HCT and cellular therapy recipients.

Introduction:

Hematopoietic cell transplant (HCT) can cure several hematologic malignancies and bone marrow failure syndromes, but the associated immunodeficiency and delayed immune reconstitution cause infectious complications. The risk of infection, type of pathogen, and timing depend on several factors, including, but not limited to, the type of HCT, choice of donor, degree of immune suppression, and development of graft versus host disease (GVHD) [1, 2]. Infectious complications in the immunocompromised host can cause significant morbidity and mortality and pose serious threats to HCT recipients [3–7]. More recently, immune effector cell therapy (IEC), also known as cellular therapy, has become an important intervention for treating several hematological malignancies [8–12]. These interventions also carry a significant infection risk due to lymphodepletion before and alteration of the immune system after the infusion [13].

The Blood and Marrow Transplant Clinical Trials Network (BMT CTN) was established in 2001 to conduct large multi-institutional clinical trials addressing important issues towards improving the outcomes of HCT and other cellular therapies. Network studies have answered critical questions about conditioning intensity, donor choice, infection prophylaxis, immune reconstitution, GVHD prevention and treatment, and relapse mitigation strategies [14–23]. To date, BMT CTN trials have enrolled more than 16,750 patients in over 55 trials. Thanks in part to studies conducted by the BMT CTN, advances in HCT have contributed to better transplant outcomes [24, 25]. As new strategies and therapies emerge and are investigated in HCT, including GVHD prevention, conditioning approaches, and cellular and gene therapies, efficacy must be balanced with the risk and severity of toxicities, notably infectious complications. Prospective grading is an important part of BMT CTN operations, with, for example, GVHD grading initiation using an algorithm published in 2003, soon after the network started [26]. “BMT-CTN Infection Grading System” is a monitoring tool developed by the BMT CTN to monitor infectious complications in the setting of these novel therapeutics, which is described in this article.

Process and Policy Development:

The Infectious Disease Technical Committee (IDTC) is an ad hoc BMT CTN committee for advising Protocol Teams, reviewing infectious disease data, and updating relevant aspects of the BMT CTN Manual of Procedures (MOP) based on published literature and changes in clinical practice. The IDTC consists of transplant physicians and infectious disease specialists with substantial experience in HCT. The IDTC recommends data collection procedures, defines infectious complications, specifies their grading, and audits data collection. A Data and Coordinating Center (DCC) Principal Investigator serves on the Committee in an ex officio capacity. There is a core group of interested, experienced investigators who are committed to the work of the Network and who convene upon request of a Protocol Team or the DCC.

The suggested detail to which infectious complications will be monitored varies from trial to trial and is decided in advance by a trial’s Protocol Team. The Protocol Team determines what, if any, variations in the medical practice of infection prevention might confound study outcomes. If there are concerns on the part of the Protocol Team, Steering Committee, or the IDTC that confounding may occur, the IDTC assists the Protocol Team in determining what data are necessary to monitor such practices or provides recommendations for standardization of those practices such that they may be collected prospectively.

Capturing Infection Adverse Events in HCT Clinical Trials and Cellular Therapy Trials:

Infectious complications are adverse events frequently observed in HCT and cellular therapy recipients and anticipated in those participating in clinical trials. Their presentation overlaps with many non-infectious syndromes, and diagnoses are complex. There are a wide array of pathogens that can cause infections in cancer patients, including bacteria, viruses, parasites, and fungi that may occur as acute infectious events such as pneumonia or bloodstream infection, or as reactivation of latent organisms as in the case of herpesvirus infections. They may also present as colonization without true invasive infection or as recurrent or resistant infections. Furthermore, sites of infections may be localized or disseminated. Such variability of clinical information can be difficult to capture accurately in a large clinical trial database.

Inconsistencies in infection reporting among immunocompromised patients enrolled in clinical trials restrict understanding of infection burden, comparability, and future pooling of data [27]. Adverse infectious events in cancer clinical trials are typically graded according to the National Cancer Institute’s (NCI) Common Terminology Criteria for Adverse Events (CTCAE) [28]. However, there are limitations to this framework as it relates to HCT and IEC trials.

CTCAE terms capture a limited number of pathogen scenarios. The CTCAE does not capture most double-stranded DNA viral infections, community respiratory viral infections, or fungal infections in sufficient detail to draw substantial conclusions that would lead to a change in clinical practice or help inform clinical implications of these infections when analyzed from the BMT CTN registry. The hierarchy of CTCAE terms includes a half dozen specific pathogen situations, and the rest are infectious syndromes without pathogen names attached. When using such a limited number of pathogen-specific terms, the consequences of infectious complications may be lost. Vice versa is also true; when reporting adverse events mainly using syndromes, the role of pathogens is lost. The CTCAE grading system does not adequately capture the complexities and nuances of infections in the context of BMT CTN trials, where patients have compromised immune systems and unique susceptibilities to infections [14]. As a result, there is a need for a more specialized grading system that considers the specific challenges and risks these patients face.

CTCAE’s emphasis on parenteral versus oral therapies during grading does not accurately reflect disease severity in patients with mucositis or GVHD, who may receive all medications parenterally for reasons unrelated to the severity of their infection. The distinction between oral and parenteral therapy is becoming outdated in an era where oral antimicrobials with high bioavailability are increasingly recognized as equally effective to intravenous medications, even for severe infections [29, 30]. CTCAE infection grading criteria do not provide detailed diagnostic criteria, methods for avoiding duplicate reporting, or techniques for capturing accurate and reproducible data. Additionally, the CTCAE criteria are not validated to predict infectious morbidity or mortality.

In the setting of IEC, the most common AE includes cytokine release syndrome, observed soon after the cell infusion. This syndrome is indistinguishable from sepsis presentation, and the CTCAE grading system does not guide navigating the overlap.

Rationale and Construction of the BMT CTN Infection Grading System:

From April 1995 to October 2000, 405 adult and pediatric patients scheduled to undergo unrelated marrow donor transplantation in 15 United States centers were randomized to receive GVHD prophylaxis with immunosuppression versus T-cell depletion [31]. Infection data were collected as part of the trial and compared between these two GVHD prophylaxis groups [32]. Infections after marrow infusion were reported by organism, body site, and severity grade, fatal, life-threatening (hypotension), severe (intravenous antibiotics), or mild/moderate (no/oral treatment). A computerized algorithm was developed at the data-coordinating center (EMMES Corporation, Rockville, MD) for a) defining disseminated infection; b) tabulating contiguous sites (Appendix B); c) defining polymicrobial infection (Appendix A of reference [32]); d) upcoding of the severity of specific infections; e) scoring infection episodes within predefined recurrence intervals (Appendix C of reference [32]); and f) censoring infections at relapse, primary or secondary graft failure with a subsequent graft infusion, or the data cut-off date of April 2002. A scheme for defining cause of death and infectious contributions to death was developed and published separately [33]. Medical records auditing at each transplant center and extensive panel review validated the computerized tabulations. The computerized scoring algorithm for infections, validated by a panel of experts, was extremely effective for data analysis from this large multicenter clinical trial [32].

In 2006, Cordonnier and colleagues prospectively recorded the type and timing of 440 infectious events in 190 consecutive patients for the initial six months post-transplant. They classified clinically significant infections into three grades according to their expected mortality rates, with grade 3 infections having a significantly increased risk of death [34]. Based on these findings, while concurrently trying to analyze infection data within the BMT CTN registry [14, 35], in 2011 the BMT CTN IDTC modified their data analysis algorithm to align with the published Cordonnier criteria to develop the BMT CTN Infection Grading System for use in BMT CTN trials. The largest modification in methodology from analysis of the T-cell depletion trial [32] when compared with the Cordonnier criteria [34] was collapse of the severity scale from 5 levels (fatal, life-threatening, severe, moderate, mild) to 3. Currently, the BMT CTN Infection Grading System guides the analysis of BMT CTN clinical trials and is designed to capture infectious complications of HCT accurately based on their potential impact on morbidity and mortality for HCT recipients.

Like the Cordonnier criteria, the BMT CTN Infection Grading System is a 3-grade system that delineates the expected clinical significance of infections caused by bacteria, viral, fungal, and parasitic organisms, as well as clinically defined infection (without identification of microbial etiology) as listed in Tables 1–7. Grade 1 infections often do not have significant impact on clinical outcomes, and some may resolve spontaneously. These infections have high rates of cure or resolution and low risk of serious sequelae. Grade 2 infections are of moderate severity, may require the administration of intravenous or oral anti-infective agents, and may require hospitalization or low-risk procedures, but generally are not lethal. Grade 3 infections are potentially life-threatening or associated with a substantial risk of morbidity. They may require prolonged or complex treatments, including surgeries and other therapies that carry risks of serious harm or require treatment in an intensive care setting. These categories reflect expert opinion on the expected clinical significance of infectious complications, and will need validation to determine whether they do, in fact, correlate with mortality and other relevant outcomes.

Table 1.

Severity Grading Table for Bacterial Infections

Type of Infection/Severity Grade	Grade 1	Grade 2	Grade 3
Bacterial infections	Bacteremia with skin flora [e.g. coagulase-negative Staphylococcus (CoNS, S. epidermidis), Corynebacterium, or Cutibacterium (Propionibacterium)] requiring antibiotics for ≤ 14 days of therapy for treatment	Bacteremia due to other organisms (not skin flora)	Bacteremia with deep organ involvement (for example, new or worsening pulmonary infiltrates, endocarditis, brain abscess) Septic Shock with bacteremia Endocarditis Brain abscess or meningitis Active tuberculosis infection
	Bacterial focus NOS requiring systemic antibiotics for ≤ 14 days of therapy for treatment (for example, urinary tract infection Bacterial focus NOS requiring only topical, ocular, or otic treatments	Bacterial focus NOS with persistent signs/symptoms or persistent positive cultures requiring antibiotics for > 14 days of therapy (GI, GU, abdominal infections, myositis)
	Cellulitis responding to initial therapy within 14 days	Cellulitis requiring a change in therapy due to progression or systemic treatment for > 14 days Localized or diffuse skin and soft tissue infections requiring incision with or without drain placement but no debridement	Fasciitis or other skin and soft tissue infection requiring surgical debridement
	Any bacterial pneumonia not requiring supplemental oxygen	Any bacterial pneumonia requiring low-flow oxygenπ	Bacterial pneumonia requiring high-flow oxygenµ or positive-pressure ventilationβ
	C. difficile toxin- or PCR-positive stool with diarrhea < 1L / day (< 5 episodes / day) without abdominal pain (child < 20 mL/kg/day)	C. difficile toxin- or PCR-positive stool with diarrhea ≥ 1 L / day (≥ 5 episodes/day) (child ≥ 20 mL/kg/day) or with abdominal pain	C. difficile toxin or PCR-positive stool with ileus, colon dilation, toxic megacolon, or need for surgical intervention

^πlow flow defined as oxygen by nasal cannula at ≤ 6L/minute

^µhigh-flow oxygen defined as oxygen by nasal cannula at >6L/minute

^βContinuous positive airway pressure (CPAP), bilevel positive airway pressure (BPAP), intubation with mechanical ventilation

Table 7:

Pediatric Systemic Inflammatory Response Syndrome Definitions and Laboratory Values Ranges for the four Pediatric Age Groups ^ϣ

Age	Tachycardia (bpm)	Bradycardia (bpm)	Tachypnea (breaths/min)	Leukocytosis / Leukopenia (103/mm3WBC)	Hypotension Systolic BP mmHg
1 mo to 1 yr	>180	<90	>34	>17.5 to <5.0	<75
>1 yr to 5 yr	>140	NA	>22	>15.5 to <6.0	<74
>5 yr to 12 yr	>130	NA	>18	>13.5 to <4.5	<83
>12 yr to < 18 yr	>110	NA	>14	>11 to <4.5	<90

^ϢGoldstein B, Giroir B, Randolph A; International Consensus Conference on Pediatric Sepsis. International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med. 2005 Jan;6(1):2–8. doi: 10.1097/01.PCC.0000149131.72248.E6. PMID: 15636651.

BMT CTN clinical trials typically record grade 2 and 3 but not grade 1 infections and maximum severity of an infection episode is captured. A standardized infection Case Report Form (CRF) is used for all protocols unless grade 1 reporting is essential to the trial endpoint. In this case, the Protocol Team designs a protocol specific CRF. This approach is consistent with FDA requirements for collecting Severe Adverse Event (SAE) data or SAEs of special interest [36]. Since these patients often are infected with multiple organisms simultaneously or sequentially identified by laboratory tests and are at risk for recurrent infections, Table 5 provides a framework to determine if an infection should be considered a new infection or a continuation of a previously identified infection.

Table 5:

Recurrence Intervals to Determine Whether An Identified Organism Reflects A Previously Diagnosed Infection or A New Infection

Type of Infection	Recurrence Internal reflects a previously diagnosed Infection
Cytomegalovirus, Herpes simplex virus, Epstein-Barr virus, and Human herpes virus 6 related infections	2 months (< 60 days)
Varicella zoster virus	2 weeks (< 14 days)
Polyomavirus	2 months (< 60 days)
Bacterial, non-C. difficile	1 week (< 7 days)
Bacterial, C. difficile	1 month (< 30 days)
Yeast infections (non-cryptococcal)	2 weeks (< 14 days)
Invasive mold infections, dimorphic fungal infection, and cryptococcal infection	3 months (< 90 days)
Helicobacter pylori infection	1 year (< 365 days)
Respiratory viruses: Adenovirus, Enterovirus, Influenza A &B, Respiratory syncytial virus, Parainfluenza, Rhinovirus, and SARS-CoV-2 infections	3 months (<90 days)
Parasitic infections (excluding chronic strongyloidiasis)	3 months (< 90 days)
Chronic strongyloidiasis (defined as positive serologies without detection of larvae)	2 years

There are three specific choices made for the BMT CTN infection grading that should be acknowledged. First, the diagnosis of a probable invasive fungal infections incorporates the decision to initiate therapy as this will help data managers identify potential clinically significant infections for intensive review. Second, the BMT CTN requires only one serum 1,3-beta-D-glucan (instead of two as required by EORTC[37]) as part of the diagnostic criteria for probable invasive Candida infection is to reflect real-world clinical practice and to facilitate retrospective identification of infections in the context of data capture for secondary endpoints in HCT clinical trials. Similarly, the BMT CTN requires only a single PCR to confirm mold infection from bronchioalveolar lavage. Though these requirements are different from the EORTC and may impact the specificity of these assays, these evidence-based decisions will avoid undercounting of clinically significant adverse events and anti-microbial related toxicity[38, 39]. Stricter criteria may be appropriate for interventional trials designed to test new treatments for infections. Second, we use the CDC sepsis reporting toolkit instead of SOFA to define adult sepsis because the CDC criteria are more readily applied retrospectively[40].

Our Viewpoint:

Adoption of new HCT and cellular therapies requires careful consideration of not only the efficacy of the treatment but also the nature and severity of its complications, which impact long-term outcomes such as non-relapse mortality and survival. Recipients of HCT incur unique infection risks compared to other populations, including the general oncology population, due to delayed immune reconstitution associated with HCT and the need for prolonged immunosuppression to prevent and treat GVHD [41]. IEC therapy recipients are prone to infectious complications due to chemotherapy, B-cell or plasma cell aplasia, associated hypogammaglobulinemia, and occasionally prolonged neutropenia [13]. Thus, infectious complications’ incidence, type, and severity must be accurately and succinctly reported. A grade 1–3 severity scale, similar to the one established by Cordonnier, has been used by the BMT CTN since 2011 to detect and concentrate on precisely identifying pertinent infection problems in these patient populations. This technique makes it simpler to record consequential infectious problems than CTCAE, which rates adverse events from 1 to 5. Additionally, this method stresses the organ site and degree of involvement combined with a specific microbiological etiology (bacterial, viral, or fungal).

Based on previous studies showing the need for infection data auditing in HCT clinical trials [32, 33, 42–44], this grading system was developed in 2011 and thereafter has been used in infection analyses of BMT CTN trials. The feasibility of this current grading system has been demonstrated in several ways. First, this system has been utilized in 14 different BMT CTN clinical trials over the past 12 years [15–23] and upcoming clinical trials (e.g., NCT05032820) . Second, data coordinators reviewed the infection data system for ease of use and helped develop training for this system. Third, this system has been implemented in clinical trials outside the BMT CTN [45, 46], making it an acceptable standard for reporting infection data for clinical trials involving HCT.

Challenges exist with the current infection grading system and its continued implementation in BMT CTN clinical trials. First, the inclusion in the IDTC of HCT and ID experts with common interests in infections in the immunocompromised host but different clinical perspectives is a strength and is necessary to maintain a widely accepted system. It’s worth considering collaborative efforts between medical experts, researchers, and relevant stakeholders to continually re-address this panel and periodically revise the infection grading system that continues to align with the evolving complexities of BMT CTN clinical trials. Second, completing periodic reviews and revisions by the IDTC is needed to ensure that this system accounts for changes in transplant practices (e.g., GVHD prophylaxis), infectious pathogens, and other risk factors. Additionally, at times data coordinators can find it challenging to extract infection data from medical records, and adjudication of infections by clinicians is necessary.

The IDTC convenes every 1–2 two years and as-needed basis to periodically reassess and update the document to have the most accurate and reliable grading system. Recent examples included changes to cytomegalovirus (CMV) severity classification following the FDA approval of letermovir for the prevention of CMV infection [47], an update of grading to include COVID-19 infection following the unprecedented recent pandemic, and incorporation of the current EORTC/MSG criteria for probable and proven fungal infections [37]. Finally, any updates to the system made by the IDTC, including changes in microorganism nomenclature, new or updated consensus criteria, and changes in severity grading, need to be overviewed by BMT CTN leadership and data coordinators to ensure ease of application and accurate and reproducible data capture by the research team. The categorization and severity grading system is summarized in table format and made publicly accessible online and through publication.

Future directions include validation of this current grading system. The Cordonnier grading system correlated infection severity with mortality in HCT recipients. Given the rich infection data repository obtained across multiple BMT CTN clinical trials, it is now possible to validate if grade 2 or 3 infections correlate with survival outcomes to improve the accuracy of the infection grading system. It is also possible to compare the BMT CTN infection grading system with the CTCAE grading system for accuracy or feasibility by retrospective and prospective studies. While such an endeavor would be resource-intensive, it would benefit future HCT and IEC clinical trial design and reproducibility. Additionally other groups have proposed collection of additional data points[27] and the technical committee is planning on evaluating this as part of its validation work.

Conclusion:

In summary, the BMT CTN infection grading system represents a highly utilized standard for reporting severe infections associated with HCT. The infection grading system continues to evolve as practice changes, with review by experts in HCT and infectious disease. Validation and correlation studies between severe infection and mortality are needed to confirm the BMT CTN infection grading system as the accepted standard and to guide further changes and improvements to this system for future application to HCT and cellular therapy clinical trials.

Table 2:

Severity Grading Table for Fungal Infections

(Definitions of proven and probable fungal infections derived from modified EORTC/MSG criteria-supplemental Table 1, possible fungal infection category omitted )

Type of Infection/Severity Grade	Grade 1	Grade 2	Grade 3
Fungal infections	Mucocutaneous candidiasis (excluding esophagitis), including oral thrush and vaginal candidiasis Dermatophyte infections (tinea)	Candida esophagitis diagnosed by endoscopy	Fungemia including candidemia
		Fungal sinusitis confirmed radiologically without orbital, brain, or bone involvement.	Fungal sinusitis confirmed radiologically with orbital, brain, or bone involvement
		Fungal pneumonia or pulmonary nodules (unless requiring high-flow oxygenµ or positive pressure ventilationβ)	Fungal pneumonia or pulmonary nodules requiring high-flow oxygen or positive-pressure ventilation
		Fungal skin and soft tissue infection without fungemia, involvement of other sites, or need for debridement α	Disseminated or deep-seated fungal infections (for example, CNS, visceral, or ocular fungal infections) or fungal infections requiring operative debridement or other surgery
		Pneumocystis jirovecii pneumonia (unless requiring high-flow oxygen or positive pressure ventilation)	Pneumocystis jirovecii pneumonia requiring high-flow oxygen or positive-pressure ventilation

^µhigh-flow oxygen defined as oxygen by nasal cannula at >6L/minute

^βContinuous positive airway pressure (CPAP), bilevel positive airway pressure (BPAP), intubation with mechanical ventilation

^αThis category is listed to capture locally invasive fungal infection that require treatment

e.g. IV related mucormycosis or skin and soft tissue involvement with endemic yeast etc.

Table 3:

Severity Grading Table for Viral Infections

Type of Infection/Severity Grade	Grade 1	Grade 2	Grade 3
Viral infections	Mucosal (oral, esophageal, vaginal, penile) or cutaneous HSV infection requiring oral antiviral therapy or observation	Mucosal (oral, esophageal, vaginal, penile) or cutaneous HSV infection requiring IV nutrition due to pain associated with infection or IV antiviral therapy	HSV infection with end-organ involvement (encephalitis, hepatitis, pneumonitis)
	Dermatomal herpes zoster (shingles) affecting ≤ 2 dermatomes	VZV infection involving 3 or more dermatomes	Severe VZV infection with end-organ involvement (encephalitis, hepatitis, pneumonitis, retinitis) or with organ dysfunction or severe sepsis (for example, coagulopathy)
	Asymptomatic CMV viremia not requiring treatment	CMV viremia requiring therapy or CMV viremia requiring a change in therapy due to resistant or refractory disease or persistent viremia beyond 4 weeks while on treatment	CMV end-organ involvement (e.g., lung, intestines, eye)
	EBV viremia not requiring treatment	EBV viremia requiring treatment	EBV PTLD
	Adenoviral infection not requiring treatment	Adenoviral upper respiratory infection, viremia, or symptomatic viruria requiring treatment	Adenovirus with end-organ involvement, including pneumonitis but excluding conjunctivitis and upper respiratory tract infections
	HHV-6 viremia not requiring treatment	HHV-6 infection with attributed symptoms or cytopenias requiring treatment	HHV-6 with end-organ involvement (such as encephalitis, hepatitis, pneumonitis)
	BK viremia or viruria with cystitis not requiring intervention except anti-spasmodic or pain medication	BK viremia or viruria with cystitis with clinical consequence requiring treatment, for example, continuous bladder irrigation, antiviral therapy, or procedural intervention	BK viremia or viruria with end-organ damage (for example, kidney injury)
		Enterocolitis with enteric (GI) viruses
	Symptomatic upper and lower tract respiratory virus (excludes adenovirus, but including SARS-Cov-2) not requiring supplemental oxygen	Viral pneumonia or pneumonitis (excludes adenovirus but includes SARS-CoV-2; see above) requiring low-flow oxygen	Lower tract respiratory viruses (including SARs-SoV-2) requiring high-flow oxygen or positive-pressure ventilation
	Viremia (virus not otherwise specified) not requiring therapy	Any viremia (virus not otherwise specified) requiring therapy	Any viral encephalitis, meningitis, or end-organ disease

Table 4:

Severity Grading Table for Parasitic Infections and Clinically Defined Infections

Type of Infection/Severity Grade		Grade 1	Grade 2	Grade 3
Parasitic infections		Giardiasis or other parasitic gastrointestinal infection with diarrhea ≤ 1L / day (< 5 episodes / day) (child ≤ 20 mL/kg/day)	Giardiasis or other parasitic gastrointestinal infection with diarrhea ≥ 1 L / day (5 episodes / day) (child ≥ 20 mL/kg/day) or with abdominal pain
		Chronic strongyloidiasis treated with oral ivermectin or other oral therapies		Strongyloides hyper-infection or disseminated infection
		Toxoplasma DNAemia without organ involvement resolving spontaneously (without treatment)	Toxoplasma DNAemia without organ involvement requiring treatment	CNS or another organ toxoplasmosis
Clinically defined infections	Pneumonia or bronchopneumonia not requiring supplemental oxygen		Pneumonia or bronchopneumonia requiring low-flow oxygen	Any acute pneumonia requiring high-flow oxygen or positive-pressure ventilation
	Fever with negative cultures responding to treatment within 14 days
	Clinically documented infection not requiring inpatient management¥		Sepsis without an identified organism (excluding patients receiving immune effector therapy diagnosed with cytokine release syndrome (CRS))	Septic shock without an identified organism (excluding patients receiving immune effector therapy diagnosed with CRS)
			Typhlitis without severe sepsis, ileus, or need for surgical intervention

^¥Patients should be reported as having CRS and graded as such, not infection, if they exhibit CRS symptoms or signs within the time frame expected for a given product AND the treating team prefers CRS over infection as the cause of the symptoms or signs

Table 6:

Definition of Fungal Infections Invasive fungal disease (IFD) due to yeasts, yeast-like fungi, and dimorphic fungi

IFD type		Criteria for proven IFD			Criteria for probable of IFD1
Endemic mycoses (e.g., Coccidioides, Blastomyces, Histoplasma)		At least one of these criteria:  ● Histopathology or direct microscopy of specimens obtained from an affected site showing the distinctive form of the fungus, or  ● Culture of the fungus from blood or specimens from an affected site			Clinical diagnosis (pulmonary, cutaneous, osseous, GI, and/or CNS) and initiation of treatment for endemic mycosis Plus, at least one of these criteria:  ● Histoplasma or Blastomyces antigen in urine, serum, or body fluid  ● Antibody to Coccidioides in cerebrospinal fluid  ● Two-fold rise of Coccidioides antibodies in 2 consecutive serum samples
Pneumocystis jirovecii pneumonia (PJP or PCP)		Detection of the organism microscopically in tissue, BAL fluid, or sputum using conventional or immunofluorescence staining			Clinical diagnosis of PJP (indicated by the initiation of treatment) Plus, at least one of these criteria:  ● ß-D-glucan (Fungitell®) ≥8 0 ng/L (pg/mL) from one serum sample (if other etiologies for elevated Fungitell have been excluded)  ● Detection of Pneumocystis jirovecii DNA by PCR from a respiratory tract specimen
Cryptococcal infection		At least one of these criteria:  ● Histopathologic, cytopathologic, or direct microscopic examination of a specimen obtained by needle aspiration or biopsy from a normally sterile site (other than mucous membranes) showing yeast consistent with Cryptococcus species (based on morphology or PCR)  ● Recovery of Cryptococcus by the culture of a sample obtained by a sterile procedure from a normally sterile site showing a clinical or radiological abnormality consistent with an infection  ● Blood culture with Cryptococcus  ● Positive cryptococcal antigen in cerebrospinal fluid or blood			Clinical diagnosis of cryptococcal infection (pulmonary, CNS, cutaneous, disseminated) indicated by the initiation of treatment Plus, at least one of these criteria:  ● Radiographic evidence of meningeal inflammation  ● Lesion on imaging consistent with cryptococcal disease
Candida and other yeast infection		At least one of these criteria:  ● Histopathologic, cytopathologic, or direct microscopic examination of a specimen obtained by needle aspiration or biopsy from a normally sterile site (other than mucous membranes) showing yeast  ● Recovery of yeast by the culture of a sample obtained by a sterile procedure from a normally sterile site showing a clinical or radiological abnormality consistent with an infection  ● Blood culture with yeast  ● Identification of fungal DNA by PCR combined with DNA sequencing when yeasts are seen in formalin-fixed paraffin-embedded tissue			Applies to Candida only Candidemia within the previous 2 weeks with at least one of these criteria:  ● Radiographic findings consistent with abscesses in liver, spleen, or brain  ● Meningeal enhancement  ● Progressive retinal exudates or vitreal opacities on ophthalmologic examination Plus, initiation of treatment and at least one of these criteria:  ● ß-D-glucan (Fungitell®) ≥80 ng/L (pg/mL) from one serum sample (if other etiologies for elevated Fungitell® have been excluded)  ● Positive T2Candida®
Invasive fungal disease (IFD) due to Aspergillus and other molds
Proven mold infection	At least one of these criteria:  ● Histopathologic, cytopathologic, or direct microscopic examination of a tissue specimen obtained by needle aspiration or biopsy in which hyphae or melanized yeast-like forms are seen accompanied by evidence of associated tissue damage  ● Recovery of a mold by the culture of a specimen obtained by a sterile procedure from a normally sterile site (with clinical or radiological evidence of infection), excluding BAL fluid, bronchial brush, bronchial aspirate, sinus specimens, and urine  ● Blood culture that yields a mold in the context of a compatible infection  ● Identification of fungal DNA by PCR combined with DNA sequencing when molds are seen in formalin-fixed paraffin-embedded tissue
Probable mold infection	Clinical feature Pulmonary aspergillosis and other pulmonary mold infections At least one of these patterns is seen on CT imaging:  ● Dense, well-circumscribed lesions  ● Air crescent sign  ● Cavity  ● Wedge-shaped, segmental, or lobar consolidation  ● Reverse halo sign (for molds other than Aspergillus) Aspergillus or other mold tracheobronchitis Tracheobronchial ulceration, nodule, pseudomembrane, plaque, or eschar seen on bronchoscopy Aspergillus and other mold sino-nasal disease At least one of these criteria:  ● Acute localized pain  ● Nasal ulcer with black eschar  ● Extension from the paranasal sinus across bony barriers Aspergillus and other mold CNS infection Focal lesions or meningeal enhancement on imaging		AND	Mycologic evidence  ● Aspergillus or other mold recovered by culture from sputum, BAL, bronchial brush, or aspirate  ● Microscopic detection of mold from sputum, BAL, bronchial brush, or aspirate  ● At least one of these criteria applied to Aspergillus galactomannan antigen:   ○ Single serum or plasma: ≥ 1.0   ○ BAL fluid: ≥ 1.0   ○ Single serum or plasma: ≥ 0.7 plus BAL fluid ≥ 0.8   ○ CSF: ≥ 1.0  ● At least one of these criteria applied to organism-specific PCR (e.g., Aspergillus or Mucor):   ○ Plasma, serum, or who   ○ le blood: 2 or more consecutive PCR tests positive   ○ BAL fluid: 1 or more PCR tests positive   ○ At least 1 PCR test positive in plasma, serum, or whole blood and 1 PCR test positive in BAL fluid

¹Modified from EORTC/MSG criteria

Highlights.

• Infection grading in HCT and IEC therapy recipients require special considerations

• BMT CTN Infection grading system is a pragmatic scoring tool to capture and grade complex infections

• This report details the creation and experience of the grading system utilized in BMT CTN trials