American Journal of Clinical Dermatology

https://doi.org/10.1007/s40257-018-0376-3

SHORT COMMUNICATION

Retrospective Chart Review of Cutaneous Adverse Events Associated

with Tremelimumab in 17 Patients
Macartney Welborn1 · Shelby L. Kubicki1 · Naveen Garg2 · Anisha B. Patel1,2

© Springer Nature Switzerland AG 2018

Abstract
Background and Objectives  Tremelimumab is a monoclonal human antibody that inhibits cytotoxic T-lymphocyte-associated
antigen 4, giving rise to increased T cell activation and interleukin-2 release. While this activation of the immune system
provides a mechanism to recognize and destroy cancer cells, it also leads to off-target immune-related adverse events. Ipili-
mumab is a US Food and Drug Administration-approved anti-cytotoxic T-lymphocyte-associated antigen 4 antibody, which
has a high incidence of cutaneous adverse events. While cutaneous adverse events for ipilimumab have been extensively
studied, there is a distinct lack of cutaneous adverse event data for tremelimumab.
Methods  We conducted a retrospective chart review of our institution’s electronic medical records from January 2000 to
March 2018 to characterize cutaneous adverse events induced by tremelimumab. Previous descriptions of tremelimumab
cutaneous adverse events are limited to rash and pruritus.
Results  We found 17 patients treated with tremelimumab who had cutaneous adverse events including pruritus (12/17),
eczematous dermatitis (8/17), morbilliform rash (5/17), vitiligo (2/17), xerosis (3/17), acneiform rash (2/17), and psoriasi-
form dermatitis (1/17).
Conclusions  This case series demonstrates that cutaneous adverse events seen in patients taking tremelimumab overlap with
those of ipilimumab. While there are some differences between rash characterizations of the two drugs, such as time to onset
and clearance, the sample size of this case series is too small to draw any definite conclusions. This study addresses a gap
in the descriptive knowledge on tremelimumab cutaneous adverse events and highlights the need for further large cohort
prospective studies. Awareness of expected cutaneous toxicities and how best to treat these can help patients continue on
immunotherapy regimens without delays or interruptions and give patients the best quality of life while receiving treatment.

Key Points  1 Introduction

There is a distinct lack of information on the description
and characterization of cutaneous adverse events caused
by tremelimumab
Tremelimumab cutaneous adverse events are similar to
those of ipilimumab but possibly differ in severity and
the length of time taking the drug before they appear
* Anisha B. Patel
APatel11@mdanderson.org
1 most common irAEs to diarrhea and colitis [2].
University of Texas McGovern Medical School, Houston,
TX, USA
2 carcinoma with tremelimumab (administered at 15 mg/kg),
University of Texas MD Anderson Cancer Center, 1515
Holcombe Blvd, FCT 11.5000, Unit 1452, Houston,
TX 77030‑4009, USA
Tremelimumab is a human IgG2 antibody that inhibits cyto-
toxic T-lymphocyte-associated antigen-4 (CTLA-4), generat-
ing an increase in immune response via T-cell activation and
interleukin-2 release. While this activation of the immune
system provides a mechanism to recognize and destroy can-
cer cells, it also leads to off-target immune-related adverse
events (irAEs) including diarrhea/colitis, dermatitis, hepa-
titis, and endocrinopathies [1]. A phase III clinical study
(tremelimumab administered at 10 mg/kg) revealed that 25%
of patients reported pruritus and 23% of patients reported
rash, making cutaneous adverse events (CAEs) the second
In a clinical trial treating patients with hepatocellular
a pruritic skin rash was the most common irAE affecting
65% of patients [3]. Despite the relatively high incidence

Vol.:(0123456789)
 M. Welborn et al.

of CAEs, there is a lack of descriptive data including the Table 2  Guidelines used to characterize vaguely described rashes
characterization and management of CAEs related to treme-
Rash characterization Key phrases
limumab in the literature. The purpose of our study was to
further describe and characterize CAEs encountered by Eczematous dermatitis Scaly
patients while taking tremelimumab and to compare these to Rough
the more widely used drug, ipilimumab. As tremelimumab Excoriations
is currently being studied in many active clinical trials, it is Open skin
important to characterize the different types of CAEs seen Sandpaper like
in patients taking tremelimumab. In this series, we present Cracking
17 cases of tremelimumab-induced CAEs. Rash that is worse in creases
Focal areas of erythema
with xerosis and pruritus
2 Methods Morbilliform Macular erythematous
Blanching
We used natural language processing to search our institu- Non-pruritic
tion’s electronic medical records for patients treated with a Covered large BSA
tremelimumab on-and-off protocol at our institution from BSA body surface area
January 2000 to March 2018. This was part of an institu-
tional review board-approved retrospective study and a
waiver of patient consent was obtained. We conducted a was listed under a miscellaneous category (acneiform, pso-
two-factor search for all tremelimumab drug names and riasiform) and subsequently described.
attempted to identify patients with cutaneous toxicities using
key rash terms as seen in Table 1. This method was very
sensitive and our cohort was narrowed through a careful 3 Results
chart review.
Rash characterization was performed using the derma- Our search found 17 patients with a variety of cancers
tologist’s diagnosis or, if not available, a combination of his- including melanoma (5/17), renal cell carcinoma (5/17),
tology, medical photography, and key phrases as described urothelial carcinoma (3/17), lung adenocarcinoma (2/17),
in Table 2. Many patients were not seen by a dermatologist mesothelioma (1/17), and prostate cancer (1/17) who had
or were not given a definitive diagnosis. The two main rash CAEs while taking tremelimumab. They were treated at dif-
types seen were eczematous and morbilliform, and the key fering times ranging from 2004 to 2017. Sixteen out of the
phrases we used to characterize each are listed in Table 2. 17 patients had stage IV cancer and one patient had a stage
All rashes that were reported without diagnosis or key words III diagnosis. The average patient age was 63 years with
were labeled as uncharacterizable. Any rash that was less a range of 48–76 years. Six of the 17 patients (35%) were
prevalent and specifically diagnosed in the physician’s notes receiving additional immunotherapy aside from tremeli-
mumab. Treatment response showed 76% (13/17) of patients
with tumor progression, while 18% (3/17) of patients had
Table 1  Drug name and key rash terms used to search our institu-
tion’s medical records tumor response and 6% (1/17) of patients had stable disease.
Only one patient was a complete responder taking tremeli-
Drug names Tremelimumab Anti-CTLA-4 Immune
Ticilimumab CTLA-4 blocker checkpoint
mumab therapy alone (patient no. 3). The CAEs and patient
CP-675,206 inhibitor characteristics seen in the 17 cases are summarized in Fig. 1
and Table 3.
Key rash terms Rash Eczema Nodosum The most common CAE was pruritus (12/17) followed
Lesion Bullae EN by eczema (8/17) with the average time to onset of CAE
Eruption Callus Vitiligo being 15 weeks. In general, therapy for eczema and pruri-
Dermatitis Keratosis pilaris Hand tus consisted of topical corticosteroids and oral antihista-
Blister Dry skin Foot mines. Cutaneous adverse events with tremelimumab had an
Vesicle Xerosis Feet average time to cutaneous clearance of 46 weeks. Eczema
Acne Mucositis Nail (seen in Fig. 2) had the longest average time to clearance at
Photo Panniculitis Pigment 54 weeks followed by pruritus with an average of 49 weeks.
Keratoderma Sweet Eruptive Patient no. 4 had folliculitis and candidiasis superinfections
CTLA-4 cytotoxic T-lymphocyte-associated antigen-4, EN erythema of eczema, and was managed with antibacterial and anti-
nodosum fungal creams respectively. Vitiligo (Fig. 3) was seen in two
Cutaneous Adverse Events Associated with Tremelimumab
Fig. 1  Characterization, time to onset, and length of cutaneous adverse events seen in patients treated with tremelimumab
patients, both showing tumor response and both had per-
sistent vitiligo after drug discontinuation. Other cutaneous
toxicities seen were morbilliform eruption (5/17), xerosis
(3/17), acneiform eruptions (2/17), and psoriasiform derma-
titis (1/17). Forty-one percent (7/17) of patients were seen by
a dermatologist for their cutaneous toxicities. One out of the
17 (6%) patients had treatment discontinuation as a result of
severe rash and pruritus, and this patient was not seen by a
dermatologist (patient no. 2).
4 Discussion
Tremelimumab is an immune checkpoint inhibitor that tar-
gets CTLA-4 and induces T-cell proliferation to increase
autoimmune tumor response. Tremelimumab and ipili-
mumab are both anti-CTLA-4 inhibitors but tremelimumab
is an IgG2 antibody and ipilimumab is an IgG1 antibody. An
IgG2 antibody is advantageous because it has reduced FcγR
binding and less complement activation as compared with
IgG1 antibodies [4]. Reduced FcγR binding could decrease
antibody-dependent cell-mediated cytotoxicity, and reduced
complement activation could result in less complement-
dependent cytotoxicity. It is also theorized that lower levels
of FcγR binding could decrease the risk of cytokine-release
syndrome, which occurs with antibody infusion in humans
[5]. Because of these properties, tremelimumab theoretically
should have fewer irAEs than ipilimumab but may also be
less effective as a cancer treatment [6, 7].
Tremelimumab is not yet approved by the US Food and
Drug Administration. In a phase III clinical trial for patients
with advanced melanoma, it was found that tremelimumab
had no statistically significant difference in tumor response
as compared to standard-of care therapy [1]. One explanation
for this may be that ipilimumab became widely accessible
to the control group while the study was ongoing (via clini-
cal trials as well as through a worldwide expanded-access
program). At least 14% of the patients in the control group
reported that ipilimumab was used and it was believed that
probably more patients had ipilimumab as part of their treat-
ment [1]. However, tremelimumab is still a highly studied
drug as there are 94 active phase I, II, and III trials currently
being conducted for various solid organ malignancies [8, 9].
Our case series presents 17 patients from our institution’s
database who had a CAE while taking tremelimumab. We
qualitatively compared the types of CAEs seen in these 17
patients taking tremelimumab to CAEs seen in patients
treated with ipilimumab in our institution’s database. Both
the tremelimumab and ipilimumab patient groups had vari-
ous cancer types and were treated from 2004 to 2017. In
both cohorts, a minority of patients were seen by a derma-
tologist [35% (6/17) of patients treated with tremelimumab
and 29% (91/310) of patients treated with ipilimumab]. As
such, consistent methods as described above to character-
ize rash type were used to further diagnose rashes for both
CTLA-4 inhibitors. Additionally, both cohorts had patients
that were receiving dual immunotherapy [35% (6/17) of
patients treated with tremelimumab and 46% (145/310) of


Table 3  Summary of cutaneous immune-related adverse events in patients treated with tremelimumab

No. Age (y), sex/race Cancer Dose (length of txs) Treatment year Seen by a Histology Therapies Tumor outcome Notes
dermatologist
(Y/N)

1 52, M/W Stage IV melanoma 10 mg/kg q4 wk (16 2004 N Right axilla; spon- Oral antihistamines Tumor progression
wk) giotic dermatitis
with eosinophils
2 54, M/W Stage III melanoma 10 mg/kg q4 wk (8 2004 N Left abdomen; spon- Oral antihistamines Tumor response Stopped tx because of
wk) giotic dermatitis rash and pruritus
with eosinophils
3 75, F/W Stage IV melanoma C1: 15 mg/kg q3 mo 2003–2009 Y None Oral antihistamines, Tumor response Complete responder
(30 wk) anesthetic cream
C2: 15 mg/kg q2 mo
(29 wk)
C3: 15 mg/kg q3 mo
(132 wk)
4 70, M/W Stage IV melanoma C1: 10 mg/kg q4 wk 2004–2011 Y Left abdomen; Topical corticoster- Tumor response
(120 wk) superficial perivas- oids, oral antihista-
C2: 10 mg/kg q3 mo cular and intersti- mines, emollient
(247 wk) tial dermatitis with
eosinophils.
Left forearm;
prurigo nodularis,
inflamed
5 76, M/W Stage IV mesothe- 10 mg/kg q4 wk (48 2014–2015 N None Oral antihistamines Tumor progression
lioma wk) and topical corti-
costeroids
6 67, M/W Stage IV urothelial 10 mg/kg q4 wk (4 2016 N None None Tumor progression
cell carcinoma wk)
7 68, M/W Stage IV lung 1 mg/kg q4 wk (25 2015–2016 N None None Tumor progression Also taking dur-
adenocarcinoma wk) valumab
8 50, M/A Stage IV lung 3 mg/kg q4 wk (5 2015 Y None Topical corticoster- Tumor progression Also taking mogam-
adenocarcinoma wk) oids ulizumab
9 69, F/W Stage IV urothelial 1 mg/kg q4 wk (22 2015–2016 Y Left elbow; skin Oral antihistamines Tumor progression Also taking dur-
cell carcinoma wk) with ulcer and and topical corti- valumab
overlying inflam- costeroids
matory scale
crust, impetigi-
nized. Associ-
ated perivascular
dermatitis with
eosinophils
10 71, M/H Stage IV renal cell 10 mg/kg q4 wk (48 2015–2016 Y None Topical corticoster- Stable response Also taking mogam-
carcinoma wk) oids ulizumab
M. Welborn et al.
Table 3  (continued)
No. Age (y), sex/race Cancer Dose (length of txs) Treatment year Seen by a Histology Therapies Tumor outcome Notes
dermatologist
(Y/N)

11 58, F/W Stage IV melanoma 10 mg/kg q4 wk (8 2004 N None None Tumor progression
wk)
12 74, M/W Stage IV renal cell 10 mg/kg q4 wk (13 2016–2017 N None Topical corticoster- Tumor progression
carcinoma wk) oids
13 52, M/W Stage IV renal cell 10 mg/kg q4 wk (4 2016 N None Topical corticoster- Tumor progression
carcinoma wk) oids
Oral antihistamines
14 70, F/W Stage IV urothelial 1 mg/kg q4 wk (24 2015–2016 Y Left elbow; skin Topical corticos- Tumor progression Also taking dur-
cell carcinoma wk) with ulcer and teroids and oral valumab
overlying inflam- antihistamines
matory crust,
impetiginized.
Cutaneous Adverse Events Associated with Tremelimumab

Associated
perivascular
dermatitis with
eosinophils
15 75, M/W Stage IV prostate 75 mg q4 wk (8 wk) 2017 Y None Topical corticos- Tumor progression Also taking dur-
cancer teroids and oral valumab
antihistamines
16 48, M/W Stage IV sarcama- 10 mg/kg (1 d/wk) 2017 N Left lateral back; Oral corticosteroids Tumor progression
toid renal cell epidermal spongi-
carcinoma osis
17 50, F/W Stage IV renal clear 10 mg/kg q4 wk (4 2017 N None Oral corticosteroids Tumor progression
cell carcinoma wk)

A Asian, C cycle, H Hispanic, N no, q every, Tx treatment, W white, Y yes


Fig. 2  a Patient no. 2, abdomen: erythematous eczematous papules coalescing into plaques. b Patient no. 2, hematoxylin and eosin stain, magni-
fication ×200: spongiotic dermatitis with eosinophils
Fig. 3  Patient no. 2, right arm: depigmented macules and patches
patients treated with ipilimumab]. These patients were not
excluded as this more closely reflects real-life experience.
The two main rash types seen with tremelimumab treat-
ment were morbilliform and eczematous dermatitis. In our
institution’s database, 21.6% (67/310) of patients taking
ipilimumab had morbilliform rash and 26.5% (82/310) had
eczema. Among our 17 patients treated with tremelimumab,
29% (5/17) had morbilliform eruptions while 47% (8/17) had
eczematous dermatitis. Differences in treatment interruption
were seen with 21% (64/310) of patients taking ipilimumab
in our institution’s database having to discontinue or inter-
rupt treatment secondary to CAEs, while one of the 17 (6%)
patients taking tremelimumab had treatment cessation.
Finally, tremelimumab CAEs appeared to have a late
median time to onset at 15 weeks while ipilimumab irAEs
have a median time to onset of 4 weeks with CAEs origi-
nating earlier in treatment than other irAEs [10]. As there
are only 17 patients being reported in this case series, no
definite conclusions on the comparison between ipilimumab
M. Welborn et al.
and tremelimumab CAEs can be drawn. Additionally, as not
all patients were seen by a dermatologist, the quantitative
values assigned to the types of CAEs could vary but there is
still benefit in understanding the qualitative data on the dif-
ferent types of CAEs seen with both drugs. These 17 cases
are a starting platform to help us understand the possible dif-
ferences and similarities between the CAEs of the two drugs.
Cutaneous adverse events to anti-CTLA-4 immune
checkpoint inhibitors could be explained by an ipilimumab/
tremelimumab-induced T-cell loss of tolerance to melano-
cyte differentiation antigens, which are prominent in both
the skin and the tumor [11]. T-cell loss of tolerance can
lead to melanocyte destruction and consequently vitiligo.
This theory is supported by vitiligo being a CAE caused by
both ipilimumab and tremelimumab. Two of five patients
with melanoma in this report developed vitiligo while taking
tremelimumab or soon after. Patients can develop vitiligo
spontaneously with melanoma or while receiving immu-
notherapy [12]. Vitiligo has been reported to be a clinical
indicator for a decreased risk of disease progression (two-
fold decrease) and death (four-fold decrease) as compared
to patients without vitiligo while receiving immunotherapy
[13]. Our case series could support this decreased risk in
that both of the patients who developed vitiligo had a tumor
response and one of the patients was a complete responder
on tremelimumab.
Six of the 17 patients (35%) were treated with co-immu-
notherapies. Two of the patients were taking mogamuli-
zumab, a human antibody chemokine receptor 4 inhibitor
reported to be associated with rash (occurring in 63% of
patients in a phase II trial) [14]. The other four patients were
taking durvalumab, which is a human antibody directed
against programmed cell death-1 ligand 1. Durvalumab was
Cutaneous Adverse Events Associated with Tremelimumab
used in a phase I trial with tremelimumab where rash was
reported in 11% of patients but there was no delineation as
to which drug caused the rash [15]. Owing to the fact that
rash could be attributed to either drug in these six patients,
it is unclear if the CAEs were due to tremelimumab alone.
This reflects the real-life experience of how patients will
be seen to evaluate and manage toxicities, as many patients
are treated with multi-drug immunotherapies. As larger
cohorts become available, co-therapy groups will be fur-
ther analyzed.
For a physician, there is a large variety of cutaneous tox-
icities to be aware of when treating a patient taking tremeli-
mumab. This range shows the importance of rash histology
for characterization and management.
Limitations to our study include low patient number and
that 35% of our patients were treated concurrently with other
immunotherapies. The inherent limitations of a retrospective
chart review exist in that rash characterization is depend-
ent on the skill level of the evaluator. We did attempt to be
consistent in our categorization as is seen in Table 2. Fur-
ther, the institutional database is limited to patients who had
CAEs; therefore, overall incidence numbers are not avail-
able. As tremelimumab is being more widely used, we will
have better information to further compare the two drugs.
5 Conclusions
Based on this limited study, tremelimumab has similar CAEs
to ipilimumab but with a possibly slower time to onset and
lower severity; however, no definite conclusions can be
made. Studies of irAEs in larger patient cohorts treated with
tremelimumab could reveal an equally efficacious immune
checkpoint inhibitor with a lower toxicity profile.
Acknowledgements  The authors thank Matthew Welborn, PhD for his
help with writing the computer code to create the graph.
Compliance with Ethical Standards  Gupta A, Narwal R, Steele K, Gu Y, Karakunnel JJ, Rizvi BA.
Funding  No external funding was received for the preparation of this
article.
Conflict of interest  Macartney Welborn, Shelby L. Kubicki, Naveen
Garg, and Anisha B. Patel have no conflicts of interest directly relevant
to the contents of this article.
References
1. Ribas A, Kefford R, Marshall MA, et al. Phase III randomized
clinical trial comparing tremelimumab with standard-of-care
chemotherapy in patients with advanced melanoma. J Clin Oncol.
2013;31(5):616–22.
2. Tremelimumab. Drugs R D. 2010;10(2):123–32. https​://doi.
org/10.2165/11584​530-00000​0000-00000​.
3. Sangro B, Gomez-Martin C, de la Mata M, et  al. A clinical
trial of CTLA-4 blockade with tremelimumab in patients with
hepatocellular carcinoma and chronic hepatitis C. J Hepatol.
2013;59(1):81–8.
4. Comin-Anduix B, Escuin-Ordinas H, Ibarrondo FJ. Tremeli-
mumab: research and clinical development. Onco Targets Ther.
2016;9:1767–76.
5. Ribas A. Clinical development of the anti-CTLA-4 antibody
tremelimumab. Semin Oncol. 2010;37(5):450–4.
6. Arce vargas F, Furness AJS, Litchfield K, et al. Fc effector func-
tion contributes to the activity of human anti-CTLA-4 antibodies.
Cancer Cell. 2018;33(4):649–63.e4.
7. He M, Chai Y, Qi J, et al. Remarkably similar CTLA-4 binding
properties of therapeutic ipilimumab and tremelimumab antibod-
ies. Oncotarget. 2017;8(40):67129–39.
8. Clinicaltrials.gov. “Clinical trials involving tremelimumab.”
13 February, 2018. http://www.clini ​ c altr ​ i als.gov/ct2/resul​
ts?term=tremel​ imuma​ b&Search​ =Apply& ​ recrs=
​ a&recrs=​ d&age_
v=&gndr=&type=&rslt=. Accessed 18 Jul 2018.
9. Joshua AM, Monzon JG, Mihalcioiu C, Hogg D, Smylie M, Cheng
T. A phase 2 study of tremelimumab in patients with advanced
uveal melanoma. Melanoma Res. 2015;25(4):342–7.
10. Lacouture ME, Wolchok JD, Yosipovitch G, Kähler KC, Busam
KJ, Hauschild A. Ipilimumab in patients with cancer and the man-
agement of dermatologic adverse events. J Am Acad Dermatol.
2014;71(1):161–9.
11. Hodi FS, Mihm MC, Soiffer RJ, et al. Biologic activity of cyto-
toxic T lymphocyte-associated antigen 4 antibody blockade in pre-
viously vaccinated metastatic melanoma and ovarian carcinoma
patients. Proc Natl Acad Sci USA. 2003;100(8):4712–7.
12. Byrne KT, Turk MJ. New perspectives on the role of vitiligo in
immune responses to melanoma. Oncotarget. 2011;2(9):684–94.
13. Teulings HE, Limpens J, Jansen SN, et al. Vitiligo-like depigmen-
tation in patients with stage III–IV melanoma receiving immuno-
therapy and its association with survival: a systematic review and
meta-analysis. J Clin Oncol. 2015;33(7):773–81.
14. Subramaniam JM, Whiteside G, Mckeage K, Croxtall JC. Moga-
mulizumab: first global approval. Drugs. 2012;72(9):1293–8.
15. Antonia S, Goldberg SB, Balmanoukian A, Chaft JE, Sanborn RE,
Safety and antitumour activity of durvalumab plus tremelimumab
in non-small cell lung cancer: a multicentre, phase 1b study. Lan-
cet Oncol. 2016;17(3):299–308.