Association of IL-6 and TNF-α with mortality in hospitalized cancer patients.

Abstract

Background:

Severe cutaneous adverse reactions (SCARs) are associated with high morbidity and mortality in cancer patients. Early identification and treatment of SCARs may improve outcomes.

Objective:

To identify biomarkers to predict outcomes in hospitalized cancer patients who developed SCARs.

Methods:

Retrospective review of 144 hospitalized cancer patients with a morbilliform rash, recorded testing for serum cytokines (IL-6, IL-10, and TNF-α) or elafin, and a dermatology consultation. Rashes were categorized as ‘simple’ morbilliform rash without systemic involvement or ‘complex’ morbilliform rash with systemic involvement.

Results:

Fifty-four of 144 (37.5%) patients died during follow-up. Elevated levels of IL-6, IL-10, and TNF-α were associated with decreased survival. OS in patients with elevated levels of IL-6, IL-10, and TNF-α was 53.7%, 56.6%, 53.6%, respectively compared to 85.7%, 82.5% and 83.6% with lower levels. Patients with increased levels of both IL-6 and TNF-α had a nearly 6-fold increase in mortality (HR 5.82) compared to patients with lower levels.

Limitations:

Retrospective design, limited sample size, and high-risk population.

Conclusions:

Hospitalized cancer patients with rash and elevated IL-6 and TNF-α were nearly 6 times more likely to die over the course of follow-up. These biomarkers may serve as prognostic biomarkers and therapeutic targets for this high-risk population.

Capsule Summary

• Reliable prognostic biomarkers are needed to predict outcomes in hospitalized cancer patients with rash.

• Hospitalized cancer patients with rash and elevated IL-6 and TNF-α levels have a nearly 6 times decreased overall survival. Given the availability of TNF-α and IL-6 inhibitors, these biomarkers may serve as potential therapeutic targets.

Introduction

Patients with cancer are at risk of developing therapy-related morbilliform eruptions, graft-versus-host disease (GVHD), and severe cutaneous adverse reactions (SCARs), including Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), and drug-induced hypersensitivity (DIHS)/drug reaction with eosinophilia and systemic symptoms (DRESS). SCARs are 28–57 times more common in cancer patients, with 5.7–14.9 cases per 100,000 patients, and more fatal (mortality of 32% with vs. 8.5% without cancer in SJS/TEN) with the greatest risk among hematologic cancer patients^1–3. The increased mortality in this population is thought to be multifactorial due to malnutrition, cancer type, immunocompromised status, and chemotherapy type.⁴

Early recognition and treatment of SCARs is vital for improving survival.⁴ Earlier withdrawal of the causative drug is associated with better SCAR prognosis (odds ratio, 0.69 for each day; 95% CI: 0.53–0.89).⁵ Given the increased incidence and mortality of SCARs in cancer patients, and the difficulty in clinically diagnosing SCARs, reliable, objective markers are needed. Our previous study found an association between elafin, interleukin-6 (IL-6), and tumor necrosis factor alpha (TNF-α) with all-cause mortality in cancer patients who develop SCARs and interleukin-10 (IL-10) and IL-6 with a drug-related complex rash.⁶ Elafin is overexpressed in wound healing and inflammatory skin disorders and is a diagnostic and prognostic plasma biomarker in GVHD. Elevated IL-6 and IL-10 levels have been found in patients with acute GVHD, SJS, TEN.^{7, 8}

Biomarkers have potential prognostic and diagnostic utility in a hospitalized cancer population who develop SCARs. The objective of this study was to identify serologic markers or combinations of markers that may be used to predict outcomes in hospitalized cancer patients who developed SCARs.

Methods

This was a retrospective cohort study approved by the Institutional Review Board of Memorial Sloan Kettering Cancer Center (MSKCC). A database query of adult patients with cancer who were hospitalized between 8/2016–2/2019 and had ICD 9 or 10 codes for rash (R23, R21, 693, 692, 695, 690–698, L20-L30, L51, L43.2, T88.7, L55–59), recorded testing for serum biomarkers IL-1β, IL-6, IL-10, TNF-α, or plasma elafin, and prior dermatology consultation revealed 151 eligible patients (Figure 1). These biomarkers are obtained at our institution as standard of care for all patients who present with possible drug eruption to complement the clinical picture and serve as a potential therapeutic target. Seven patients were excluded because biomarkers were checked for a reason other than a morbilliform rash (i.e., cytokine release syndrome, study protocol, sepsis, or cellulitis/panniculitis).

Figure 1:

Flowchart of patient selection. GVHD, Graft-versus-host disease; ICD, International Classification of Diseases; MSKCC, Memorial Sloan Kettering Cancer Center; 2/2, secondary to

One hundred and forty-four patients were evaluated on the inpatient service or in the urgent care center at MSKCC with a diagnosis of morbilliform rash and tested for cytokines or elafin. Chart review was performed for all patients to assign ‘simple’ and ‘complex’ morbilliform rash groups. ‘Simple’ morbilliform rash was defined as a rash with no systemic involvement, or spontaneous resolution of rash with remote systemic involvement, or limited course of rash that did not require systemic therapy. ‘Complex’ morbilliform rash was defined as a SCAR with systemic organ involvement requiring systemic therapy with a prolonged course.

For each patient, a modified RegiSCAR score⁹ was calculated based on the following items: fever ≥ 38.5°C (0 points, −1 if absent); peripheral eosinophilia (≥ 700/mm³ or ≥ 10%; 1 point or ≥ 1500/mm³ or ≥ 20%; 2 points); presence of atypical lymphocytes (1 point); rash ≥ 50% of body surface area (1 point) with facial edema, purpura, infiltration, or desquamation (1 point); organ involvement (1 point for one organ, 2 points for two or more); disease duration > 15 days (0 points, −1 if absent); at least 3 biological investigations (e.g., blood cultures, viral serology, biopsy) performed and negative to rule out an alternative diagnosis (1 point). Comprehensive metabolic panel, including glomerular filtration rate (GFR), blood urea nitrogen (BUN), creatinine (Cr), transaminases, total bilirubin and urine eosinophils, were reviewed. For all laboratory values, only results within 7 days of biomarker testing were recorded to minimize the impact of non-rash events on biomarker levels. Reference values for cytokines (IL-10 ≤18 pg/mL, IL-6 ≤5 pg/mL, TNF-α ≤22 pg/mL) are determined by our institution’s laboratory, while that for elafin (≤23.8 ng/mL) is determined by Viracor, Inc.

Descriptive statistics and graphical methods were used to assess distributions of patient and medical test characteristics. Chi-square tests and Fisher’s Exact test were used to assess the association between rash type (simple versus complex) and nominally scaled patient and medical test characteristics. T-tests and Wilcoxon rank-sum tests were used to assess differences in continuously scaled variables by rash type. Overall survival (OS) was calculated as the time elapsed between initial assessment for SCARs to death from any cause. A patient was considered censored at the date of last follow-up date if she or he were alive at that point of contact. Kaplan-Meier estimates with log rank tests were used to assess probabilities of survival. Cox proportional hazards regression models were used to explore prognostic factors associated with survival while controlling for patient characteristics. Proportionality assumptions for Cox models were evaluated by the visual assessment of Cox-Snell and standardized residuals. All p-values are two-sided. Analyses were performed with STATA 16.0 software (StataCorp LP, College Station, TX).

Results

Patient characteristics

One hundred and forty-four patients with cancer and morbilliform rash who were evaluated at a single institution for dermatology consultation were included in the study. Of these patients, 50.6% were male, and the average age was 55.5 years. Of the 144 patients, 81 (56.3%) had a ‘simple’ rash and 63 (43.7%) had a ‘complex’ rash with systemic involvement. Among complex rash patients, 46% (n=29) had cutaneous graft-versus-host disease (GVHD), 6.3% (n=4) had cutaneous GVHD-spectrum rash (engraftment syndrome), and the remaining 46.9% (n=30) had a complex rash secondary to drug exposure (Table I). Most patients with complex rash had hematologic malignancy (69.8%), while most patients with a simple rash had solid malignancy (56.8%).

Table I.

Characteristics of hospitalized cancer patients and ‘simple’ morbilliform rash vs. ‘complex’ morbilliform rash. GVHD, Graft-versus-host disease; BSA, Body surface area; CTCAE, Common Terminology Criteria for Adverse Events; GFR, glomerular filtration rate; SCAR, severe cutaneous adverse reaction.

Category		Rash type		Total	p-value
		Simple morbilliform rash, n=81, n (%)	Complex morbilliform rash, n=63), n (%)
Patient Sex	Female	40 (49.4)	31 (49.2)	71 (49.3)	0.983
	Male	41 (50.6)	32 (50.8)	73 (50.7)
Location	Inpatient	49 (60.5)	53 (84.1)	102 (70.8)	0.002
	Urgent care center	32 (39.5)	10 (15.9)	42 (29.2)
Cancer type	Solid	46 (56.8)	18 (28.6)	64 (44.4)	0.001
	Hematologic	32 (39.5)	44 (69.8)	76 (52.8)
	Both	3 (3.7)	1 (1.6)	4 (2.8)
Diagnosis	complex drug	1 (1.2)	30 (47.6)	31 (21.5)	<0.001
	GVHD	0 (0)	29 (46)	29 (20.1)
	simple drug	77 (95.1)	0 (0)	77 (53.5)
	GVHD-like	0 (0)	4 (6.4)	4 (2.8)
	viral exanthem	3 (3.7)	0 (0)	3 (2.1)
CTCAE Grade	0	0 (0)	1 (1.6)	1 (0.7)	0.002
	1	9 (11.1)	0 (0)	9 (6.3)
	2	28 (34.6)	13 (20.6)	41 (28.5)
	3	44 (54.3)	46 (73)	90 (62.5)
	4	0 (0)	3 (4.8)	3 (2.1)
RegiSCAR score	0	33 (40.7)	3 (4.8)	36 (25)	<0.001
	1	28 (34.6)	14 (22.2)	42 (29.2)
	2	16 (19.8)	14 (22.2)	30 (20.8)
	3	3 (3.7)	24 (38.1)	27 (18.8)
	4	1 (1.2)	7 (11.1)	8 (5.6)
	5	0 (0)	1 (1.6)	1 (0.7)
Atypical	No	78 (96.3)	54 (87.1)	132 (92.3)	0.041
Lymphocytes	Yes	3 (3.7)	8 (12.9)	11 (7.7)
Rash, (50% BSA + purpura, edema, scale)	0	37 (46.3)	3 (4.8)	40 (28)	<0.001
	1	30 (37.5)	29 (46)	59 (41.3)
	2	13 (16.3)	31 (49.2)	44 (30.8)
Fever, >38°C	No	5 (6.2)	8 (12.7)	13 (9)	0.175
	Yes	76 (93.8)	55 (87.3)	131 (91)
Decreased GFR relative to baseline	No	72 (90)	47 (74.6)	119 (83.2)	0.04
	Yes	8 (10)	16 (25.4)	24 (16.8)
Elevated transaminases relative to baseline	No	68 (84)	28 (44.4)	96 (66.7)	<0.001
	Yes	13 (16.1)	35 (55.6)	48 (33.3)
Elevated total bilirubin relative to baseline	No	76 (93.8)	55 (87.3)	131 (91)	0.175
	Yes	5 (6.2)	8 (12.7)	13 (9)
Internal organs involved	0	59 (72.8)	12 (19.1)	71 (49.3)	<0.001
	1	22 (27.2)	43 (68.3)	65 (45.1)
	2	0 (0)	8 (12.7)	8 (5.6)
Respiratory virus panel positive	No	43 (86)	46 (93.9)	89 (89.9)	0.193
	Yes	7 (14)	3 (6.1)	10 (10.1)
Human herpesvirus positive	No	36 (87.8)	41 (82)	77 (84.6)	0.445
	Yes	5 (12.2)	9 (18)	14 (15.4)
Cytomegalovirus positive	No	42 (93.3)	48 (90.6)	90 (91.8)	0.618
	Yes	3 (6.7)	5 (9.4)	8 (8.2)
Epstein-Barr virus positive	No	41 (93.2)	52 (92.9)	93 (93)	0.950
	Yes	3 (6.8)	4 (7.1)	7 (7)
Adenovirus positive	No	40 (100)	48 (98)	88 (98.9)	0.364
	Yes	0 (0)	1 (2)	1 (1.1)
Skin biopsy supportive of drug reaction	No	56 (69.1)	32 (50.8)	88 (61.1)	0.025
	Yes	25 (30.9)	31 (49.2)	56 (38.9)
Status	Alive	53 (65.4)	35 (55.6)	88 (61.1)	0.228
	Dead	28 (34.6)	28 (44.4)	56 (38.9)

Biomarkers

The median values of cytokines (IL-10, IL-6, TNF-α) and elafin were higher in the complex rash patients compared to those with simple rash, although only IL-10 reached statistical significance (p=0.03) (Table II). Peripheral eosinophilia and white blood cell count did not differ between simple and complex rash groups.

Table II.

Comparison of median values of patient characteristics by ‘simple’ morbilliform rash vs. ‘complex’ morbilliform rash. IL, interleukin; TNF-α, tumor necrosis factor α

Characteristic	Overall Median	Rash type	Median	Binomial Interpolation of 95% CI		p-value
				Lower	Upper
Age (years)	60.5	Simple	62	56.2	66.8	0.17
		Complex	56	53.0	63.0
IL-1b	0	Simple	0	0	0	0.38
		Complex	0	0	0
IL-10	17.5	Simple	14	10.0	19.5	0.03
		Complex	26.5	16.0	31.6
IL-6	24.5	Simple	17.5	10.5	30.0	0.14
		Complex	31.5	22.8	50.8
TNF-α	14	Simple	12.5	10.0	16.5	0.27
		Complex	16	13.4	19.0
Elafin	18.35	Simple	17	13.3	23.3	0.43
		Complex	20.2	17.2	28.0
% Eosinophils	1.95	Simple	2.1	1.0	2.8	0.80
		Complex	1.8	0.2	4.7
Absolute eosinophils	0.1	Simple	0.1	0.1	0.2	0.96
		Complex	0.15	0	0.3
White blood cells	5.65	Simple	5.5	4.6	6.7	0.72
		Complex	5.7	4.3	7.8

Biomarkers and organ involvement

The median IL-10 level was significantly higher in patients with elevated transaminases than in patients with transaminases in the normal range (24.5 pg/mL vs. 14 pg/mL, p= 0.01). Median IL-6 levels were also elevated in patients with elevated transaminases, however this difference did not reach statistical significance (41 pg/mL vs 19.5 pg/mL, p=0.056). There was no significant association between IL-10, IL-6, TNF-α, or elafin and bilirubin or renal involvement, as measured by decreased glomerular filtration rate relative to baseline.

Overall survival

Median follow-up time for the cohort was 14.7 months. Fifty-four of 144 (37.5%) patients died during follow-up. Six-month OS was not statistically significantly different between complex versus simple rash, or between patients with a Common Terminology Criteria for Adverse Events (CTCAE) v5.0 score of 0–2 versus 3–4. Patients with morbilliform rash and elevated IL-6 levels had a decreased OS that worsened in concordance with rising IL-6 levels. At 6 months, patients with IL-6 levels above the median (>24 pg/mL) had an OS of 53.7% (95% CI: 39.7–65.5), compared to 85.7% OS (95% CI: 73.5–92.6) in those with lower IL-6 levels (≤ 24pg/mL), p<0.001. Patients with TNF-α levels above the median value (>14 pg/mL) had a significantly shorter OS at 6 months than those with TNF-α levels in the bottom two quartiles (≤14 pg/mL), 56.6% versus 82.5%, p<0.001. Elevated IL-10 levels were also associated with decreased 6-month OS, as patients with IL-10 levels greater than the median (≥18 pg/ml) had a 6-month survival of 53.6% (95% CI: 42.5–68.0) compared to 83.6% (95% CI: 70.9– 91.1) for those below the median, p=0.006. Table III presents univariate and multivariate Cox models for OS. In univariate analysis, higher levels of IL-6, IL-10 and TNF-α were all associated with worse prognosis. On multivariate analysis examining cytokines, rash type (simple versus complex) and patient age, patients with elevated levels of IL-6 were 3.2 times more likely to die (HR = 3.21, 95% CI: 1.64–6.32) over the course of the follow-up compared to those with lower levels. Similarly, patients with higher levels of TNF-α were also more likely to die compared to those with lower levels (HR = 1.78, 95% CI 0.94– 3.35). Figure 2 depicts the multivariate survival experience of patients with the combination of increased levels of IL-6 and TNF-α. These patients had a 5-fold increase in mortality compared to those with lower levels of both measures (HR =5.82, 95% CI: 1.53– 17.43). No difference was observed in 6-month overall survival in cancer patients with simple versus complex rash and elevated absolute eosinophils.

Table III.

Univariate and multivariate analyses of Overall Survival using Cox proportional hazards models. IL, interleukin; TNF-α, tumor necrosis factor α

Category		Univariate			Multivariate*
		HR for death (95% CI)	p-value		HR for death (95% CI)	p-value
Rash Type	Simple	referent	0.165		referent	0.666
	Complex	1.46 (0.85– 2.51)			0.88 (0.48– 1.59)
IL-6	0–24	referent	<0.001		referent	0.001
	25/max	3.77 (2.02–7.04)			3.21 (1.64– 6.32)
IL-10	min/17	referent	0.008		referent	0.386
	18/max	2.22 (1.23– 4.02)			1.35 (0.68– 2.70)
TNF-α	min/14	referent	0.001		referent	0.076
	15/max	2.66 (1.46– 4.84)			1.78 (0.94– 3.35)

^*Variables included in multivariate are the ones listed in the table, along with age.

Figure 2.

Survival functions for patients with IL-6 and TNF-α above vs. below the median. In both groups, the model predictions are based on patients with a mean age of 55.5 years. Variables included in the multivariate analyses are IL-6, TNF-α, IL-10, rash type (simple morbilliform vs complex morbilliform), and age. IL, interleukin; TNF-α, tumor necrosis factor α

Discussion

This study examined patients with cancer who presented with morbilliform rash and found the combination of elevated TNF-α and IL-6 to be significantly associated with a nearly 6 times increased all-cause mortality after rash onset (p<0.001). While past studies have shown an increased mortality associated with elevated IL-10, IL-6, and elafin, the prognostic value of combinations of cytokines have not been elucidated.^{6, 10} Using multiple biomarkers as opposed to single cytokines can strengthen the overall prognostic value and serve as a reliable tool for SCAR diagnosis and prognostication.¹¹ Our study did not find a statistically significant difference in OS at 6 months between patients with complex and simple rashes (69.5% compared to 75.8%, respectively, p=0.23), highlighting the need for objective tools to guide patients who require therapy. The availability of TNF-α and IL-6 inhibitors make these biomarkers attractive therapeutic targets for intervention. The Food and Drug Administration’s approval of the IL-6 receptor antibody tocilizumab for the treatment of cytokine release syndrome (CRS) further underscores the therapeutic potential of identifying inflammatory markers in supportive onco-dermatology.¹² The severity of symptoms in CRS may correlate with duration of exposure to the inflammatory cytokines, emphasizing the importance of prompt recognition of these markers for both diagnostic and treatment purposes.¹³

Our study found that levels of IL-6 in cancer patients with morbilliform rash are inversely related to OS, confirming in a larger cohort the previous report by Mori et al.⁶ IL-6 is a pleiotropic cytokine with pro-inflammatory effects that inhibit the induction of regulatory T-cells by TGF-beta and shifts naïve T-cell differentiation towards Th17 cells.^{14, 15} Th17 cells release large quantities of IL-17, IL-22, and IL-21, and is a major inducer of tissue inflammation.¹⁶ Significant elevations in serum IL-6 have previously been reported in patients with SCARs and have also been associated with increased incidence and severity of GVHD.^{6, 11, 17} IL-6 is implicated in the development of rashes and elevated lesional IL-6 levels have been found in numerous other inflammatory skin disorders .^18–20 Administration of recombinant IL-6 to a patient resulted in an inflammatory skin eruption with recurrence after rechallenge.²¹ Although the exact mechanism is not well understood, the rapid resolution of rash in CRS with the use of IL-6 inhibitors implies the pathophysiologic importance of IL-6.¹² Moreover, IL-6 mediates tumor progression and therapeutic resistance.^22–24 Elevated levels of IL-6 in cancer patients have been associated with poor outcomes and addition of an IL-6 blockade to immune checkpoint inhibitor therapy and mRNA-based immunotherapy has resulted in a significant reduction in tumor volume and led to prolonged survival in mice.^{22, 25} IL-6 blockade reversed anti-PD-L1 resistance and prolonged tumor-bearing mouse survival in a preclinical colorectal cancer model.²⁶ IL-6 blockade also augments regulatory T cell reconstitution.¹⁷ IL-6 inhibitors have been effectively used in GVHD prophylaxis and treatment of steroid-refractory acute and chronic GVHD.^27–29 IL-6 blockade receptors in murine models of GVHD significantly reduced GVHD-associated mortality.³⁰ Given our findings of decreased OS in cancer patients with rash and high IL-6, IL-6 may represent an actionable therapeutic target to improve OS in this patient population.

TNF-α, a pro-inflammatory cytokine, appears to be involved in the pathogenesis of severe cutaneous adverse reactions. TNF-α induces keratinocyte apoptosis through a Fas-mediated pathway, leading to the extensive epithelial cell death seen in SJS/TEN.³¹ This mechanism of action may explain the elevated TNF-α frequently found in the plasma and blister fluids of patients with SJS/TEN, and in the plasma of patients with GVHD and acute generalized exanthematous pustulosis (AGEP).^{7, 32–35} Additionally, TNF-α has been linked to promotion and progression of cancer.^{36, 37} Levels of IL-6 and TNF-α directly correlated with the extent of disease in patients with prostate cancer.³⁸ TNF-α was also an independent predictor of poor survival in cancer patients, highlighting the potential role of TNF-α antagonists in the treatment of cancer patients.^{39, 40} Furthermore, TNF-α inhibitors are an effective treatment for steroid-refractory aGVHD and SCARs.^{32, 41–43} Moreover, a randomized controlled trial found that TNF-α antagonists decrease the predicted mortality rate and healing time in patients with cytotoxic T lymphocyte-mediated SCARs (including SJS-TEN).⁴⁴

In our study, elevated IL-10 levels in cancer patients with morbilliform rash were significantly associated with subsequent development of a complex rash (p=0.03) and decreased OS at 6 months (p=0.006). IL-10 is an anti-inflammatory cytokine produced by Th2 cells and keratinocytes in the skin that inhibits the production of IL-6 and TNF-α.^{45, 46} IL-10 may act as a defense mechanism against excessive tissue inflammation.⁴⁷ Patients with SJS/TEN may have elevated blister fluid concentrations of IL-10.⁷Past studies have similarly found that elevated IL-10 levels may be produced as a result of excessive tissue inflammation in toxic epidermal necrolysis and reflect a compensatory response.⁴⁶ Elevated levels of IL-10 have also been correlated with a poor response in numerous types of cancer, however the role of IL-10 in tumorigenesis remains controversial.^48–50 Its immunosuppressive influence has been thought to reduce the anti-tumor immune response, yet murine tumor models have demonstrated rapid tumor rejection with increased IL-10 secretion.⁵¹ Given the significantly decreased 6-month OS (p=0.006) in patients with elevated IL-10 levels in our cohort, our study supports prior literature in cancer patients and suggests that this marker may also have prognostic value in cancer patients at the time of rash development.^{49, 52}

Although eosinophilia is a criterion on the RegiSCAR scoring system for identifying patients with DRESS/DIHS, there was no difference in 6-month overall survival in cancer patients with morbilliform rash and elevated absolute eosinophils. Past studies have similarly found that eosinophils are associated with immune-related adverse events but not overall survival.⁵³

Limitations of this study include its retrospective design and limited sample size. All cases were recruited from a tertiary referral cancer center. A larger, prospective study examining the effect of IL-6 or TNF-α inhibitors on overall survival in patient with elevated IL-6 or TNF-α is needed. This exploratory analysis presents potential therapeutic targets to improve overall survival in a high-risk patient population.

Conclusion

This study highlights the prognostic significance of biomarker combinations for cancer patients with a morbilliform rash. The combination of elevated IL-6 and TNF-α in this population is significantly associated with a nearly 6 times decreased OS (P<0.001), while IL-10 may predict progression of a simple morbilliform rash to a SCAR. A growing number of studies suggest an increased survival associated with IL-6 and TNF-α inhibition in cancer or SCARs. Further studies evaluating the impact of TNF-α and IL-6 inhibitors on survival in this population are needed.

Abbreviations

SCAR

severe cutaneous adverse reaction

SJS

Stevens-Johnson syndrome

TEN

toxic epidermal necrolysis

DIHS

drug-induced hypersensitivity syndrome

DRESS

drug reaction with eosinophilia and systemic syndrome

GVHD

graft versus host disease

IL-6

interleukin-6

IL-10

interleukin-10

TNF-α

tumor necrosis factor alpha

GFR

glomerular filtration rate

OS

Overall Survival