The n e w e ng l a n d j o u r na l of m e dic i n e

disease, and mild hypertension), but we believe
that this finding is misleading. The problem is
that the single-nephron GFR was calculated di-
rectly from the number of nephrons (specifically,
the total GFR divided by the number of neph-
rons), and thus, the measurement error with the
number of nephrons will alone result in a nega-
tive correlation between the single-nephron GFR
and the number of nephrons. A sophisticated
statistical simulation with ideally more measure-
ment-error data than we had may help clarify the
biologic correlation.
In general, we agree with Rosenberg and
Hostetter that glomerular enlargement might be
considered harmful. We have found that aging
alone does not lead to glomerular hypertrophy,
despite a considerable decline in the number of
nephrons. However, certain risk factors for
chronic kidney disease, such as obesity and fam-
ily history of end-stage renal disease, are associ-
ated with glomerular enlargement and an ele-
vated single-nephron GFR. Nonetheless, kidney
donation itself leads to glomerular enlargement
with increased single-nephron GFR in the remain-
ing kidney but without evidence of glomerular
hypertension,2 and progressive chronic kidney
disease rarely occurs after kidney donation.
There may be a tipping point at which glomeru-
lar enlargement becomes maladaptive. Atubular
glomeruli were not assessed in our study, because
we analyzed only two consecutive 3-m kidney-
biopsy sections, which limited our ability to
assess the entire glomerulus. Atubular glomeruli
are often ischemic-appearing,3 and ischemic-
appearing glomeruli account for only 0.6% of
the nonsclerosed glomeruli in this healthy popu-
lation.4
As discussed by Zarogiannis et al., the single-
nephron GFR may be higher in the deep (juxta-
medullary) glomeruli than in the more superfi-
cial glomeruli in humans, as has been observed
in studies in animals. Estimates of the single-
nephron GFR cannot be calculated separately for
deep and superficial nephrons in humans by the
method we used. Since varying the dietary so-
dium intake in healthy humans does not change
the total GFR,5 it is also unlikely that the mean
single-nephron GFR varies according to the level
of sodium intake.
Aleksandar Denic, M.D., Ph.D.
Mayo Clinic
Rochester, MN
RichardJ. Glassock, M.D.
David Geffen School of Medicine at the University of Califor-
nia, Los Angeles
Los Angeles, CA
AndrewD. Rule, M.D.
Mayo Clinic
Rochester, MN
rule.andrew@mayo.edu
Since publication of their article, the authors report no fur-
ther potential conflict of interest.
1. Hughson M, Farris AB III, Douglas-Denton R, Hoy WE, Ber-
tram JF. Glomerular number and size in autopsy kidneys: the
relationship to birth weight. Kidney Int 2003;63:2113-22.
2. Lenihan CR, Busque S, Derby G, Blouch K, Myers BD, Tan JC.
Longitudinal study of living kidney donor glomerular dynamics
after nephrectomy. J Clin Invest 2015;125:1311-8.
3. Marcussen N. Atubular glomeruli in renal artery stenosis.
Lab Invest 1991;65:558-65.
4. Kremers WK, Denic A, Lieske JC, et al. Distinguishing age-
related from disease-related glomerulosclerosis on kidney biopsy:
the Aging Kidney Anatomy study. Nephrol Dial Transplant 2015;
30:2034-9.
5. Mallamaci F, Leonardis D, Bellizzi V, Zoccali C. Does high
salt intake cause hyperfiltration in patients with essential hyper-
tension? J Hum Hypertens 1996;10:157-61.
DOI: 10.1056/NEJMc1709128

Glucocorticoid Sparing of Benralizumab in Asthma

To the Editor: Nair et al. (June 22 issue)1 found zge ztrkAkta, M.D.
in the ZONDA trial that the median blood eosino- BlentE. ekerel, M.D.
phil counts fell dramatically after treatment with A.Fuat Kalyoncu, M.D.
benralizumab. In a previous trial, Bleecker et al.2 Hacettepe University
found that blood eosinophil counts were reduced Ankara, Turkey
by benralizumab treatment. To avoid bias, what bsekerel@hacettepe.edu.tr
was done to prevent unmasking of the trial- No potential conflict of interest relevant to this letter was re-
ported.
group assignment to the investigators by means
of the patients eosinophil counts? 1. Nair P, Wenzel S, Rabe KF, et al. Oral glucocorticoidspar-

1204 n engl j med 377;12nejm.org September 21, 2017

The New England Journal of Medicine

Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright 2017 Massachusetts Medical Society. All rights reserved.
 Correspondence

ing effect of benralizumab in severe asthma. N Engl J Med 2017; ment, the investigator was instructed to avoid
376:2448-58.
ordering a complete blood cell count with dif-
2. Bleecker ER, FitzGerald JM, Chanez P, et al. Efficacy and
safety of benralizumab for patients with severe asthma uncon- ferential counts. Third, each trial center employed
trolled with high-dosage inhaled corticosteroids and long-acting a person who did not have direct interaction with
2-agonists (SIROCCO): a randomised, multicentre, placebo-
the patient to screen and redact any eosinophil,
controlled phase 3 trial. Lancet 2016;388:2115-27.
basophil, or monocyte results that were included
DOI: 10.1056/NEJMc1709523
as part of laboratory assessment reports ordered
outside the trial. The measures that were under-
To the Editor: Nair and colleagues found that taken to prevent unblinding are provided in Sec-
benralizumab showed significant, clinically rele- tion 6.6 of the trial protocol, which is available
vant benefits, as compared with placebo, on oral with the full text of our article at NEJM.org.
glucocorticoid use and exacerbation rates. How- Cao and colleagues raise an important ques-
ever, certain proportions of patients who smoke tion, but we do not believe that the data obtained
were enrolled in the benralizumab group and the in the ZONDA trial shed more light on the rela-
control group. Since the biologic effects of gluco- tionship between smoking and response to ben-
corticoids are attenuated in persons with asthma ralizumab in patients receiving oral glucocorti-
who smoke,1 can the authors report the efficacy coid therapy, since persons who were active
of benralizumab in patients with asthma who smokers and patients with a history of heavy
smoke? smoking (10 pack-years) were excluded from
Chao Cao, Ph.D. the trial. A total of 46 of 220 patients (21%) in
Xue Kong, M.D. the trial were former smokers (29 of whom re-
Xiaoping Huang, M.B. ceived benralizumab and 17 placebo). Besides
Ningbo First Hospital smoking status at trial entry and total pack-
Ningbo, China years, no other information regarding nicotine
xiaopinghuang@126.com
use was collected. The small sample size and
No potential conflict of interest relevant to this letter was re-
ported. lack of data regarding when these patients
stopped smoking preclude a meaningful analy-
1. Chaudhuri R, Livingston E, McMahon AD, Thomson L, Bor- sis of the effect of benralizumab on glucocorti-
land W, Thomson NC. Cigarette smoking impairs the thera-
peutic response to oral corticosteroids in chronic asthma. Am J coid sparing in these patients. However, on the
Respir Crit Care Med 2003;168:1308-11. basis of previously published studies of benraliz
DOI: 10.1056/NEJMc1709523 umab1 and mepolizumab,2 the anti-eosinophilic
effects of antiinterleukin-5 and antiinterleukin-
The authors reply: In reply to ztrk Akta 5-receptor therapies are unlikely to be affected
and colleagues: we expected that patients receiv- by a history of tobacco smoking.
ing benralizumab would have substantially lower Parameswaran Nair, M.D., Ph.D.
blood eosinophil counts than would patients re- McMaster University
ceiving placebo. Therefore, we took precautions Hamilton, ON, Canada
to mitigate unblinding for the trial investigators parames@mcmaster.ca
and staff who were directly involved in the care Peter Barker, Ph.D.
of the patients. First, with the exception of the Mitchell Goldman, M.D.
blood eosinophil counts that were obtained by AstraZeneca
the local laboratory at screening, all subsequent Gaithersburg, MD
Since publication of their article, the authors report no fur-
hematologic assessments were conducted by a ther potential conflict of interest.
central laboratory. Any eosinophil, basophil, or
monocyte counts were redacted from the labora- 1. Brightling CE, Bleecker ER, Panettieri RA Jr, et al. Benraliz
umab for chronic obstructive pulmonary disease and sputum
tory reports that were returned to the investiga- eosinophilia: a randomised, double-blind, placebo-controlled,
tors. Second, trial investigators were requested to phase 2a study. Lancet Respir Med 2014;2:891-901.
be mindful of ordering local laboratory tests that 2. Dasgupta A, Kjarsgaard M, Capaldi D, et al. A pilot ran-
domised clinical trial ofmepolizumab in COPD with eosino-
might inadvertently provide information regard- philic bronchitis. Eur Respir J 2017;49(3):1602486.
ing the patients blood cell counts. For example, DOI: 10.1056/NEJMc1709523
if an investigator required a hemoglobin assess- Correspondence Copyright 2017 Massachusetts Medical Society.

n engl j med 377;12nejm.org September 21, 2017 1205

The New England Journal of Medicine
Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright 2017 Massachusetts Medical Society. All rights reserved.