[Skip to Content]
[Skip to Content Landing]
Views 4,034
Citations 0
Invited Commentary
September 10, 2020

Immune Stimulation With Recombinant Human Granulocyte Colony–Stimulating Factor for Coronavirus Disease 2019 (COVID-19)—Beware of Blind Spots

Author Affiliations
  • 1Division of Pulmonary, Allergy, and Critical Care Medicine and Center for Translational Lung Biology, University of Pennsylvania Perelman School of Medicine, Philadelphia
  • 2Institute for Immunology, University of Pennsylvania Perelman School of Medicine, Philadelphia
  • 3Department of Anesthesiology and Critical Care, Children's Hospital of Philadelphia, University of Pennsylvania Perelman School of Medicine, Philadelphia
  • 4Department of Systems Pharmacology and Translational Therapeutics, University of Pennsylvania Perelman School of Medicine, Philadelphia
  • 5Parker Institute for Cancer Immunotherapy at the University of Pennsylvania Perelman School of Medicine, Philadelphia
JAMA Intern Med. Published online September 10, 2020. doi:10.1001/jamainternmed.2020.5536

The sobering mortality of coronavirus disease 2019 (COVID-19), the pandemic triggered by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and the few proven treatments have stimulated a global effort to rapidly discover and disseminate novel therapies. A noteworthy feature of the disease is the marked degree of lymphopenia in many hospitalized patients with COVID-19 and the association of lymphopenia with adverse outcomes.1

Lymphopenia during infection, and particularly during sepsis, is not unique to COVID-19, nor even to viral illness. In studies of hospitalized patients with positive blood cultures, most had lymphopenia, and persistent lymphopenia was associated with mortality.2 However, during many acute viral respiratory infections, lymphopenia is transient and coincident with peak symptoms but then rapidly resolves as the patient improves.3 The severity, and in some cases persistence, of lymphopenia in patients with COVID-19 is different.3 As patients with sepsis have concomitant inflammation and features of immune exhaustion, an attractive strategy might augment specific immune responses, particularly if it could be predicted which patients would be most likely to benefit.

In this issue of JAMA Internal Medicine, Cheng et al4 report an open-label randomized trial of recombinant human granulocyte colony-stimulating factor (rhG-CSF) compared with usual care in 200 hospitalized adults with COVID-19, pneumonia, and lymphopenia (absolute lymphocyte count, ≤ 800/μL [to convert to ×109/μL, multiply by 0.001]).4 Patients with preexisting conditions, baseline leukocytosis, and those requiring invasive mechanical ventilation at the time of screening were excluded. The patients in the treatment arm received 3 daily doses of subcutaneous rhG-CSF, 5 μg/kg. Usual care included oxygen, assisted ventilation if needed, and antibiotics or adjuvant therapies (corticosteroids, lopinavir-ritonavir, arbidol, or inhaled α-interferon) at the discretion of the treating physician.

The primary outcome was the time to clinical improvement of at least 1 point on a 7-point ordinal scale of clinical and respiratory severity ranging from not hospitalized with normal activities to death.4 The treatment groups were reasonably well balanced at enrollment, with a slightly more participants in the usual care arm receiving high-flow oxygen and adjuvant therapies before randomization. There was no difference in the primary outcome by treatment group. Treatment with rhG-CSF increased lymphocyte and leukocyte cell counts by day 5. Participants randomized to receive rhG-CSF were less likely to progress to invasive ventilation or shock and exhibited lower 21-day mortality. Both of these secondary end points were statistically significant.

Cheng et al4 tested for heterogeneity of rhG-CSF treatment effect by patient factors, including degree of lymphopenia, degree of baseline oxygen support, age, and sex. They detected significant interaction between the effect of rhG-CSF and the degree of lymphopenia, stratifying at a lymphocyte cell count of 400 or less per μL, with the participants with more severe lymphopenia exhibiting greater improvement. In contrast, for the roughly 50% of the study population with a lymphocyte cell count more than 400 per μL, no effect of rhG-CSF was observed. Similarly, Cheng et al4 report a significant interaction between the level of oxygen support and the rhG-CSF effect. Participants requiring high-flow oxygen or noninvasive ventilation at enrollment demonstrated a significant improvement in the primary outcome. In contrast, those requiring low-flow oxygen or no supplemental oxygen did not manifest a benefit.

Although the subgroup analyses suggesting a heterogeneous effect of rhG-CSF on COVID-19 recovery are provocative, caution should be exercised in interpreting these data, as the findings could be biased. For example, the stratification threshold for lymphopenia (≤400/μL) is not validated. These and other stratified results should be viewed as hypothesis-generating; they require prospective testing before they should influence clinical care. Also, it is not clear whether the administration of adjuvant therapies without a proven benefit for patients with COVID-19 affected the trial results. The more frequent use of these agents in the usual care arm suggests that clinicians viewed patients in this group as having greater illness severity or higher risk, perhaps explaining the higher proportion with worse outcomes.

Conducting a randomized interventional trial during a pandemic is no small feat, and the current trial is a substantial accomplishment. Cheng et al4 explicitly enriched the study population for patients they hypothesized would benefit in requiring lymphopenia to be eligible. Lymphocyte cell counts are an attractive enrichment marker as an objective measure that is clinically available. Prior trials of rhG-CSF for patients with pneumonia and sepsis showed no benefit, although these trials enrolled predominantly patients with bacterial pneumonia and without leukopenia.5

As a treatment choice, rhG-CSF is somewhat unconventional. First, G-CSF has a major effect on the myeloid compartment, significantly increasing neutrophil proliferation, differentiation, mobilization, and survival. Myeloid cells, including neutrophils, have been associated with more severe disease from COVID-19.1 In a nonrandomized cohort study, blockade of the granulocyte-macrophage–CSF receptor reportedly benefited patients with COVID-19 with systemic hyperinflammation.6 For these reasons, the authors excluded patients with leukocytosis.

Second, the effect of rhG-CSF on lymphocyte quantity and differentiation is somewhat limited. As reported, rhG-CSF approximately doubled the lymphocyte cell count compared with a 4-fold increase in the total leukocyte cell count.4 The G-CSF receptor is only moderately expressed in the lymphoid lineage, predominantly on developing thymocytes. One possibility is that G-CSF treatment stimulates the emergence of these T cell precursors from the thymus into the periphery, although the rapidity of the observed effect casts some doubt on whether this is the case. Another possibility is that a small subset of peripheral mature T cells responds to rhG-CSF, or that rhG-CSF–stimulated myeloid lineage cells interact with the T cell compartment to improve its fitness. Beyond stimulation, G-CSF also has regulatory roles and might downregulate neutrophil cytokine production or reset emergency hematopoiesis. Therefore, it is unclear if the effects of rhG-CSF reported by Cheng et al4 are attributable to an increase in new lymphocytes, changes to activation status of mature lymphocytes, or changes centered on a myeloid to lymphocyte axis of cell survival.

Specifically reversing sepsis-associated lymphopenia has been the focus of immunoadjuvant trials of recombinant human interleukin-7 (rhIL-7), a cytokine favoring lymphocyte survival and proliferation. In a small phase IIb trial of patients with septic shock, rhIL-7 increased lymphocyte cell counts and was well tolerated, although no effect on mortality was seen.7 RhIL-7 has also been administered to a small series of 12 critically ill patients with COVID-19; the lymphocyte cell count increased, although changes in inflammatory cytokines or survival were not seen.8

Recent work with deep immune profiling of peripheral lymphocytes has also highlighted the dramatic heterogeneity of immune responses to COVID-19, even among hospitalized patients with lymphopenia.9,10 Substantial variation in the degree of T cell and B cell activation and proliferation was evident, with organ failure associating with specific CD8 lymphopenia yet a highly active T cell response, along with a sizable and persistent accumulation of plasmablasts, antibody-secreting cells that are typically low in abundance and only transiently detectable in blood.9 Innate immune dysregulation was also evident, with decreased expression of the FcγRIII receptor (CD16) on multiple cell types that might indicate either an activated or refractory state.10 How rhG-CSF treatment, with a likely more dominant impact on myeloid cells, influences lymphocyte responses remains to be determined. Future studies should investigate whether rhG-CSF treatment has distinct effects in patients with different baseline immunological features.

The trial of rhG-CSF conducted by Cheng et al4 was relatively small, excluded patients with comorbidities, and was conducted early during the COVID-19 pandemic. Moreover, since this trial was conducted, the standard of care for hospitalized patients with COVID-19 has evolved considerably to include the use of remdesivir and dexamethasone. It is unclear how receipt of these medications in a uniform manner might influence the effects of rhG-CSF. Important next steps in determining the role of rhG-CSF or other medications that target lymphopenia in patients with COVID-19 are larger trials that incorporate the evolving standard of care into the control arm and the inclusion of patients with comorbidities.

Back to top
Article Information

Corresponding Author: Nuala J. Meyer, MD, MS, Hospital of the University of Pennsylvania, Pulmonary, Allergy, and Critical Care Medicine Division, 3600 Spruce St, 5038 Gates Building, Philadelphia, PA 19104 (nuala.meyer@pennmedicine.upenn.edu.)

Published Online: September 10, 2020. doi:10.1001/jamainternmed.2020.5536

Conflict of Interest Disclosures: Dr. Wherry reported personal fees and being a founder of Arsenal Bioscience; personal fees from Merck, Roche, Pieris, and Elstar; personal fees from and serving as scientific advisor for Surface Oncology; and holding a patent on the PD-1 pathway licensed to Roche/Genentech. No other disclosures were reported.

References
1.
Liang  W, Liang  H, Ou  L,  et al; China Medical Treatment Expert Group for COVID-19.  Development and validation of a clinical risk score to predict the occurrence of critical illness in hospitalized patients with COVID-19.   JAMA Intern Med. 2020;180(8):1-9. doi:10.1001/jamainternmed.2020.2033PubMedGoogle ScholarCrossref
2.
Drewry  AM, Samra  N, Skrupky  LP, Fuller  BM, Compton  SM, Hotchkiss  RS.  Persistent lymphopenia after diagnosis of sepsis predicts mortality.   Shock. 2014;42(5):383-391. doi:10.1097/SHK.0000000000000234PubMedGoogle ScholarCrossref
3.
Chen  Z, John Wherry  E.  T cell responses in patients with COVID-19.   Nat Rev Immunol. 2020;20(9):529-536. doi:10.1038/s41577-020-0402-6PubMedGoogle ScholarCrossref
4.
Cheng  L-l, Guan  W-j, Duan  C-y,  et al.  Effect of recombinant human granulocyte colony–stimulating factor for coronavirus disease 2019 (COVID-19) with lymphopenia: a randomized clinical trial.   JAMA Intern Med. Published online September 10, 2020. doi:10.1001/jamainternmed.2020.5503Google Scholar
5.
Root  RK, Lodato  RF, Patrick  W,  et al; Pneumonia Sepsis Study Group.  Multicenter, double-blind, placebo-controlled study of the use of filgrastim in patients hospitalized with pneumonia and severe sepsis.   Crit Care Med. 2003;31(2):367-373. doi:10.1097/01.CCM.0000048629.32625.5DPubMedGoogle ScholarCrossref
6.
De Luca  G, Cavalli  G, Campochiaro  C,  et al.  GM-CSF blockade with mavrilimumab in severe COVID-19 pneumonia and systemic hyperinflammation: a single-centre, prospective cohort study.   Lancet Rheumatol. 2020;2(8):e465-e473. doi:10.1016/S2665-9913(20)30170-3PubMedGoogle ScholarCrossref
7.
Francois  B, Jeannet  R, Daix  T,  et al.  Interleukin-7 restores lymphocytes in septic shock: the IRIS-7 randomized clinical trial.   JCI Insight. 2018;3(5):98960. doi:10.1172/jci.insight.98960PubMedGoogle ScholarCrossref
8.
Laterre  PF, François  B, Collienne  C,  et al.  Association of interleukin 7 immunotherapy with lymphocyte counts among patients with severe coronavirus disease 2019 (COVID-19).   JAMA Netw Open. 2020;3(7):e2016485-e2016485. doi:10.1001/jamanetworkopen.2020.16485PubMedGoogle ScholarCrossref
9.
Mathew  D, Giles  JR, Baxter  AE,  et al; UPenn COVID Processing Unit.  Deep immune profiling of COVID-19 patients reveals distinct immunotypes with therapeutic implications.   Science. 2020;eabc8511. doi:10.1126/science.abc8511PubMedGoogle Scholar
10.
Kuri-Cervantes  L, Pampena  MB, Meng  W,  et al.  Comprehensive mapping of immune perturbations associated with severe COVID-19.   Sci Immunol. 2020;5(49):eabd7114. doi:10.1126/sciimmunol.abd7114PubMedGoogle Scholar
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    ×