Cancer Immunol Immunother (2016) 65:37–45

DOI 10.1007/s00262-015-1773-6

ORIGINAL ARTICLE

Biomarkers related to immunosenescence: relationships

with therapy and survival in lung cancer patients
Danay Saavedra1 · Beatriz García1 · Patricia Lorenzo‑Luaces1 · Amnely González1 ·
Xitlally Popa1 · Karla P. Fuentes1 · Zaima Mazorra1 · Tania Crombet1 ·
Elia Neninger2 · Agustin Lage1 

Received: 5 December 2014 / Accepted: 6 November 2015 / Published online: 20 November 2015
© Springer-Verlag Berlin Heidelberg 2015

Abstract  Conclusions  Distribution of lymphocyte subsets was influ-

Purpose  There are well-known alterations occurring enced by cancer and chemotherapy in NSCLC patients.
within the immune system with aging. Collectively, these CD19 + B cells decrease by cancer disease and not by chemo-
changes are known as immunosenescence. The incidence therapy, and CD28− subpopulations increase by chemotherapy
of malignancies also increases with age. The aim of this and not by cancer. The proportion of CD8 + CD28− T cells,
study was to determine the presence of immunosenescence CD4+ T cells and CD4/CD8 ratio can be used as predictive
biomarkers in non-small cell lung cancer (NSCLC) patients biomarkers of CIMAvax-EGF efficacy in NSCLC patients and
and to evaluate some of them as predictive biomarkers of thereby could, be a useful tool for a personalized treatment.
CIMAvax-EGF cancer vaccine efficacy.
Methods  Sixty-six NSCLC patients, vaccinated or not Keywords  Immunosenescence · Immunosenescence
with CIMAvax-EGF cancer vaccine, and 37 age-matched markers · Non-small cell lung cancer · Cancer vaccine
controls were enrolled. Peripheral blood samples were
studied for CD19+, CD4+, CD8+, CD28−, CD57+ and Abbreviations
CD45RA+ subpopulations by flow cytometry. ANOVA Analysis of variance
Results  Absolute count of CD19+ and the CD4/CD8 BCG Bacille Calmette-Guérin
ratio were significantly lower in NSCLC patients than BSA Bovine serum albumin
in age-paired controls, while highly differentiated T cells CMV Cytomegalovirus
increased in NSCLC patients treated with platinum-based CP NSCLC patients who had not yet started first-
chemotherapy. Using Cox regression, we were able to line chemotherapy
dichotomize the patient population according to bio- CP + Ch  NSCLC patients treated with platinum-based
markers. Vaccinated patients with frequency <24 % of chemotherapy
CD8 + CD28− T cells, >40 % of CD4 T cells and CD4/ Cy5 Indodicarbocyanine
CD8 ratio higher than two at the beginning of immuno- EDTA Ethylenediaminetetraacetic acid
therapy achieved a 20-month increase in median survival EGF Epidermal growth factor
regarding control patients. FDA Food and drug administration
FITC Fluorescein isothiocyanate
Danay Saavedra and Beatriz García have contributed equally to
HR Hazard ratios
this article. HV Healthy volunteers
IRP Immune risk profile
* Danay Saavedra NSCLC Non-small cell lung cancer
danays@cim.sld.cu
OS Overall survival
1
Clinical Immunology Department, Center of Molecular PBS Phosphate-buffered saline
Immunology, 216 St, Corner 15, PO Box 16040, Havana, PE Phycoerythrin
Cuba RPE R-phycoerythrin
2
Hermanos Ameijeiras Hospital, Havana, Cuba WBC White blood cells

13

38 Cancer Immunol Immunother (2016) 65:37–45

Table 1  Demographics characteristic of patients and controls as CD28− [8]. Chen and colleagues suggested that the
Healthy volun- Cancer Cancer immune impairment in patients with cancer is associated
teers patients patients + Cha with various factors, such as the stage of cancer and the
impact of treatment [8]. Therefore, it would be enlighten-
n 37 30 36
ing to separate the effect of these factors.
Age 62.54 (4.96) 62.52 (7.93) 56.54 (7.40)
Recently, we reported the effect of aging on immune cell
(mean + SDb)
subsets in a cohort of healthy donors of all ages with high
<60 years 11 (51–59) 11 (50–59) 23 (45–59)
(range) prevalence of CMV infection. With age, the number and
>60 years 26 (60–70) 19 (60–76) 13 (60–74) frequency of B cells and T cells decreased significantly,
(range) while highly differentiated T cells increased. Such changes
Female 15 11 14 were different in males and females [9].
Male 22 19 22 A therapeutic cancer vaccine (CIMAvax-EGF), designed
a
to induce specific antibodies against the epidermal growth
  Ch chemotherapy
b
factor (EGF) [10, 11], has been developed in our institute.
  SD standard deviation
More than 3000 advanced NSCLC patients have received
CIMAvax-EGF vaccine, which has demonstrated immuno-
Introduction genicity, safety and efficacy [10].
Since the aging process entails a progressive decline
Aging of an individual is well known to be related to func- in the functional capacity of numerous organ systems and
tional limitation of immunity, resulting from age-associated loss of homeostasis [12], the search of biomarkers for prog-
changes in both the innate and the adaptive branches of the nosis and response to therapeutic vaccination should be a
immune system [1]. Those age-related alterations in the research in the field of geriatric oncology. We present here
elderly´s immune system entail an increased susceptibility the results of a prospective study providing description of
to developing infectious diseases, cancer, Alzheimer’s dis- immunosenescence biomarkers scores in Cuban cancer
ease, osteoporosis and autoimmunity [2, 3]. patients and age-matched controls. This research suggests
During the past years, the scientific community has been biomarkers to predict which patients will receive higher
interested in finding immunosenescence biomarkers, able clinical benefits with CIMAvax-EGF cancer vaccine.
to predict immune capacity and proper response to vaccina-
tion [4]. The concept of “immune risk profile” (IRP) was
defined in Sweden, based on a series of prospective cohort Materials and methods
studies of very elderly. The IRP was characterized by a
relative deficit in the number and proportion of B cells, an Subjects
accumulation of CD8+ memory T cells, CD4/CD8 ratio
less than one and an increase in late-stage differentiated Sixty-six patients with histological confirmed stage IIIB or
cells such as CD8 + CD28− T cells. Additionally, most IV NSCLC, from a phase-III clinical trial, were evaluated.
of these late-stage differentiated cells were also found to Those patients were previously randomized for receiving
express CD57 receptor [5]. an EGF-based vaccine as switch maintenance or best sup-
More than 60 % of newly diagnosed cancer patients and portive care. This study is registered in http://www.who.int/
more than 70 % of cancer-related deaths occur in subjects ictrp/network/rpcec/en/, the Cuban Public Registry of Clin-
older than 65 years [2]. As the global population ages, a ical Trial (Spanish acronym: RPCEC), a WHO-validated
rapid increase in elderly people with malignant tumors can Public Registry, Trial Number RPCEC00000161.
be expected [6]. Aged patients compose a heterogeneous Patients were subdivided considering the chemothera-
group of patients due to their comorbidities and functional peutic regime: 30 patients with NSCLC who had not
status; therefore, they represent a challenging group to the yet started first-line chemotherapy (CP) and 36 NSCLC
oncologist and scientist [7]. patients treated with platinum-based chemotherapy
Although a connection between immunodeficiency and (CP  + Ch). In addition, 37 healthy volunteers (HV) were
oncogenesis has been proposed, a detailed immune profile recruited as controls. Demographic characteristics of the
in cancer has not been well described yet. It is known that selected cohort are summarized in Table 1.
the immune system deteriorates with cancer progression as The thirty-six NSCLC patients treated with platinum-
a consequence of decreasing levels of functional T cells, based chemotherapy were randomized 30 days after com-
downregulation of the costimulatory molecules CD27 and pleting chemotherapy to receive CIMAvax-EGF vaccine
CD28 and expansion of hyporesponsive populations such or best supportive care (Fig. 1). All patient groups were
homogeneous in terms of age and sex distribution.

13
Cancer Immunol Immunother (2016) 65:37–45 39
TOTAL
n=103
NSCLC Healthy
patients Volunteers
n=66 n=37
Cancer patients Cancer
patients + Ch
n= 30
n=36
CIMAvax-EGF Control
patients patients
n=22 n=9
Fig. 1  Distribution of NSCLC patients and healthy volunteers
included in the study
The study was approved by the Institutional Review
Boards of the participating hospitals and by the National
Regulatory Authority. All patients and healthy volunteers
provided informed consent before their inclusion in the
study.
White blood cell collection and staining for flow
cytometry
Peripheral blood samples were collected from patients and
healthy volunteers by venipuncture. Red cells were lyzed
from whole blood with lysing solution (NH4Cl, EDTA
tetrasodium, KHCO3). WBCs were washed twice with
cytometry solution (PBS, BSA, Azide 20 %). Specific anti-
bodies against CD8 (FITC and PE), CD28 (PE), CD45RA
(RPE-Cy5), CD4 (FITC), CD57 (FITC) and CD3 (R-phy-
coerythrin-indodicarbocyanine (RPE-Cy5)) (AbD Serotec)
were used for staining in the following panels: CD19; CD4
FITC/CD45RA RPE-Cy5/CD28 PE; CD8/CD45RA RPE-
Cy5/CD28 PE; and CD3 RPE-Cy5/CD8 PE/CD57 FITC.
For surface staining, WBCs (1 × 105 cells in 100 μL
of cytometry solution) were incubated in the dark at 4 °C
for 20 min with 5 μL (0.25 μg) of the antibody. Subse-
quently, cells were washed twice and analyzed on a three-
color FACScan flow cytometer gating on the total lymphoid
populations. To obtain the absolute numbers, we used total
lymphocyte count from clinical laboratory analysis. The
absolute number of cells per milliliter expressing a par-
ticular phenotype was calculated multiplying lymphocyte
count by the frequency of different subsets in the lympho-
cyte gate. Samples were obtained and studied individually.
All data were obtained from fresh samples. Analysis was
performed using FlowJo data analysis software.
Statistical analysis
Statistical significance between cancer patient groups and
healthy volunteers was evaluated using nonparametric
Kruskal–Wallis test within one-way ANOVA or Mann–
Whitney test. Post hoc paired comparisons with Dunn’s
multiple comparison test were done for the variables that
showed differences between groups after ANOVA analysis.
Cox regression was adjusted to determine the relationship
between the immunosenescence biomarkers and survival,
in those patients who had been treated with the therapeutic
vaccine CIMAvax-EGF. The statistical analyses were per-
formed with SPSS v15.0 and GraphPad 5. The statistical
data were considered significant if p < 0.05. Cox regression
allowed the selection of cut points to identify the group that
includes the largest number of patients benefited by the
treatment. The method was implemented in R-software.
Results
B and T cells
Significantly lower absolute counts of CD19+ cells were
found in cancer patients as compared to healthy volun-
teers regardless of treatment with chemotherapy (Fig. 2a),
suggesting that the contraction of B cells is related to
the presence of cancer disease and not to chemotherapy
(p = 0.0002, ANOVA).
Cancer patients (independently of the use of chemo-
therapy) showed a decreased absolute count of CD4 T cells
(Fig. 2b, d.1–d.3) and an increased number of CD8 T cells
(Fig.  2c, d.1–d.3) although these trends did not reach sta-
tistical significance. However, CD4/CD8 ratio is signifi-
cantly lower in cancer patients as compared to age-matched
healthy volunteers (Fig. 2e) (p = 0.0317, ANOVA).
Besides, the CD4/CD8 ratio could also be analyzed as
the proportion of subjects with the ratio less than one. In
this case, cancer patients showed the highest proportion
of individuals with an inverted CD4/CD8 ratio (Fig. 2f),
although statistical significance was not achieved (Fisher
test, p > 0.05).
Differentiated T cells
Progressive stages in T cell differentiation can be identified
by sequential changes in the expression of surface recep-
tors such as CD45RA, CD28 [13] and CD57 [14]. In our
study, when we explored the expression of CD28 into CD4
and CD8 T cells, we found a significant increase in the
proportion of CD4 + CD28− T cells in patients with plat-
inum-based chemotherapy (p = 0.0129, ANOVA) (Fig. 3a,
13

40 Cancer Immunol Immunother (2016) 65:37–45

(a)
Absolute count (x109 cells/L)
CD19+ B cells
(b) (c)
***
0.3 CD4+ T cells CD8+ T cells
***
1.0 0.8

Absolute count

Absolute count
0.8

(x109 cells/L)

(x109 cells/L)
0.2 0.6
0.6
0.4
0.4
0.1
0.2 0.2

0.0 0.0
0.0
HV CP CP+Ch HV CP CP+Ch
HV CP CP+Ch

(d1) (d2)

(d3) (e) (f)

CD4/CD8 ratio lower than unity
4 *
CD4/CD8 ratio

25.71
3 30.00
21.43
25.00

Proportion
2 20.00
15.00 8.11
1 10.00
5.00
0 0.00
HV CP CP+Ch HV CP CP+ Ch

Fig. 2  Absolute counts of CD19 B cells, CD4 and CD8 T cells
in patients and controls. a B cell populations decreased in cancer
patients (p  = 0.0002, ANOVA, Dunn test). b The absolute count of
CD4+ (c) and CD8+ T cells do not change in cancer patients or in
cancer patients with platinum-based chemotherapy. d.1 Flow cytom-
etry profile of expression of CD4 and CD8 T cells in HV. d.2 Flow
cytometry profile of expression of CD4 and CD8 T cells in CP.
d.3 Flow cytometry profile of expression of CD4 and CD8 T cells
in CP + Ch. e The CD4/CD8 ratio decreased in cancer patients
(p  = 0.0317, ANOVA, Dunn Test). f NSCLC patients showed
the highest proportion of CD4/CD8 ratio lower than unity (CD4/
CD8 < 1), although statistical significance was not achieved (Fisher
test), “Proportion”: The frequency of patients and healthy individu-
als with CD4/CD8 ratio under one. The asterisks indicate statisti-
cally significant differences among the groups (p < 0.05). HV healthy
volunteers, CP NSCLC patients who had not yet started first-line
chemotherapy, CP + Ch NSCLC patients treated with platinum-based
chemotherapy

c–e) compared to healthy volunteers and to cancer patients with limited capacity for self-renewal [13, 15]. We exam-
without chemotherapy. ined the proportion of these subpopulations in cancer
A similar pattern of rising CD28− cells among these patients and healthy subjects. No differences were found
three groups was observed for CD8 T cell subpopulation between cancer patients and healthy volunteers (data not
(Fig.  3b, c–e) (p  = 0.0263, ANOVA). The proportion of shown).
CD8 + CD28− T cells was significantly lower in healthy In addition, we measured the expression of CD57, a
volunteers and in cancer patients without chemotherapy. marker of terminal differentiation on human CD8+ T cells
However, cancer patients treated with platinum-based because of its association with poor responses to prolifera-
chemotherapy show the highest proportions of this CD8 T tive stimuli and with replicative senescence [16]. The pro-
cell subset. portion of CD8 + CD57+ T cells did not differ among the
Several reports on CD8+ and CD4+ T cells suggest that groups (data not shown).
the CD45RA+ CD28− subset is terminally differentiated

13
Cancer Immunol Immunother (2016) 65:37–45 41

Fig. 3  Proportion of CD28− (a) CD4+CD28- T cells (b) CD8+CD28- T cells

in CD4 and CD8 T cells in 100 * 40 *
patients and controls. a The *
CD4 + CD28− (p = 0.0129, 80
30
ANOVA, Dunn test) and b

Proportion
Proportion
CD8 + CD28− (p = 0.0263, 60
ANOVA, Dunn Test) T cell 20
populations increased in 40
patients treated with chemother- 10
20
apy. c Flow cytometry profile of
expression of CD4 + CD28− T 0 0
cells and CD8 + CD28− T HV CP CP+Ch HV CP CP+Ch
cells in HV. d Flow cytom-
etry profile of expression of (c)
CD4 + CD28− T cells and
CD8 + CD28− T cells in CP.
e Flow cytometry profile of
expression of CD4 + CD28− T
cells and CD8 + CD28− T
cells in CP + Ch. The asterisks
indicate statistically significant
differences among the groups
(p < 0.05). HV healthy volun-
teers, CP NSCLC patients who
had not yet started first-line
chemotherapy, CP + Ch
NSCLC patients treated with
platinum-based chemotherapy

(d)

(e)

13

42 Cancer Immunol Immunother (2016) 65:37–45

Immunosenescence markers as predictive biomarkers
of CIMAvax‑EGF efficacy
We next explored if immunosenescence biomarkers could
be useful to predict the clinical outcome of patients treated
with a therapeutic cancer vaccine, CIMAvax-EGF.
NSCLC patients treated with chemotherapy were
recruited for a phase-III trial, testing an active immunother-
apy strategy as switch maintenance. These patients were
randomized in two groups: patients treated with CIMAvax-
EGF therapeutic vaccine (n  = 22) and control nonvacci-
nated patients (n = 9). CIMAvax-EGF is an active immu-
notherapy containing one of the most important ligands of
the EGFR coupled to a carrier protein and administered
intramuscularly together with an oil adjuvant. This thera-
peutic vaccine has proven to be immunogenic and safe in
previous clinical trials [17].
A single assessment of lymphocyte subpopulations was
performed after chemotherapy and before vaccination. We
found that vaccinated patients with higher proportion of
CD8  + CD28− T cells at the beginning of immunother-
apy have higher mortality risk (COX regression, p = 0.02)
and also vaccinated patients with higher proportion of
CD4 T cells have lower mortality risk (COX regression,
p = 0.006) (data not shown).
Based on Cox regression results, we dichotomized the
population according to the proportion of CD8 + CD28−
T cells, the proportion of CD4+ T cells and the CD4/CD8
ratio. While a modest benefit of 1.44 months in OS was
observed in the cohort of all vaccinated patients (manu-
script in preparation), noteworthy differences in OS were
detected between groups of patients according to selected
immunosenescence biomarkers, achieving a median sur-
vival superior in 20 months in comparison with control
patients. The hazard ratios were lower than 0.2 (Table 2).
Vaccinated patients with >40 % of CD4+ T cells
(50 %) achieved a median survival of 46.4 months versus
12.3 months in control patients (Fig. 4a), whereas those

Table 2  Summarized results of survival analysis

Value n % HR OS median differences Median CIMAvax-EGF Median control p value

CD4+ >40 % 18 53 0.17 34.1 46.4 12.3 0.01

CD8 + CD28− <24 % 15 55 0.20 22.9 37.2 14.3 0.03
CD4/CD8 >2 11 33 0.16 36.1 50.4 14.3 0.04

Hazard ratios, overall survival median differences and p value from vaccinated and control patients selected by the dichotomization of biomark-
ers (proportion of CD8 + CD28− T cells, proportion of CD4+ T cells or CD4/CD8 ratio, respectively). HR hazard ratios, OS overall survival

Fig. 4  Survival functions
according to T cell subpopula-
tions. a Vaccinated (n = 13)
and control (n = 3) patients
with the proportion of CD4+
T cells >40 % (p = 0.01). b
Vaccinated (n = 10) and control
(n = 6) patients with the pro-
portion of CD4+ T cells <40 %
(p > 0.05). c Vaccinated
(n = 9) and control (n = 6)
patients with the proportion of
CD8 + CD28− T cells < 24 %
(p = 0.03). d Vaccinated
(n = 9) and control (n = 3)
patients with the proportion of
CD8 + CD28− T cells < 24 %
(p > 0.05). e Vaccinated (n = 7)
and control (n = 4) patients
with the CD4/CD8 ratio >2
(p = 0.04). f Vaccinated
(n = 14) and control (n = 8)
patients with the CD4/CD8
ratio >2 (p > 0.05)

13
Cancer Immunol Immunother (2016) 65:37–45 43
with <24 % of CD8 + CD28− T cells (55 %) survived
significantly longer than control patients (median survival
37.2 vs. 14.3 months, respectively) (Fig. 4c). When com-
paring vaccinated and control patients who had the CD4/
CD8 ratio >2 (Fig. 4e), the vaccine group (33 %) achieved
a median survival of 50.4 months, whereas the control arm
had a median survival of 14.3 months.
Vaccinated patients who had the proportion of CD4+
T cells <40 % (Fig. 4b), the proportion of CD8 + CD28−
T cells >24 % (Fig. 4d) and the CD4/CD8 ratio below 2
(Fig.  4f) did not benefit from vaccination with CIMAvax-
EGF as compared to nonvaccinated control patients with
the same phenotype.
In summary, vaccinated patients with high baseline
values of CD4 T cells and CD4/CD8 ratio, as well as low
prevaccination proportion of CD8 + CD28− T cells, had
higher clinical benefit compared with control subjects.
Discussion
The risk of cancer and the senescence of the immune sys-
tem increase with aging. Both processes run at the same
time, and it is unknown in what extension they are caus-
ally related [1]. Moreover, immunosenescence is a com-
plex phenomenon including changes in different lympho-
cyte subpopulations and lymphocyte differentiation, which
could have different role in oncogenesis [2, 12, 18]. Chem-
otherapy in cancer treatment further complicates the pic-
ture, as cytotoxic drugs also affect lymphocyte populations
[8, 19].
In spite of the encouraging results of monoclonal anti-
bodies and therapeutic vaccines in cancer immunotherapy,
the relationship between immunosenescence and the prob-
ability of response to these treatments has been scarcely
addressed. Therefore, the contribution of immunosenes-
cence to the development, progression, and treatment of
cancer in elderly patients is unclear.
This article summarizes the evaluation of immunose-
nescence biomarkers in three populations: NSCLC patients
who had not yet started first-line chemotherapy, NSCLC
patients treated with platinum-based chemotherapy and a
group of age-matched healthy individuals. It also illustrates
the potential predictive value of some of these markers for
the efficacy of CIMAvax-EGF; an innovative EGF-based
therapeutic cancer vaccine.
Essentially, we found that CD19+ B cells and CD4/
CD8 ratio decrease in cancer patients but did not further
decrease after platinum-based chemotherapy, whereas
CD28− CD4+ and CD28− CD8+ T cells increase after
first-line conventional chemotherapy. In addition, we
showed no shift in CD57+ and CD45RA + CD28− T cell
subpopulations among the three groups.
Aging drives the adaptive immune system to substantial
alterations of B and T cell compartments. While age-related
changes in the B cell compartment are less known, modifi-
cations of T cell subsets have been well studied [16, 20].
Several studies describe defects in B cell generation and
decreases in class switch recombination, the process that
produces protective antibodies and memory B cells [21].
Chen and colleagues showed differences in immune
profile between cancer patients and normal subjects. They
found decreased proportions of CD4+ and CD8+ T cell
subpopulations in cancer patients, mainly at more advanced
disease stages and in those patients who had received con-
ventional chemotherapy treatment [8].
Although we did not find significant differences in the
counts of CD4 and CD8 T cells separately across the three
study groups, we found a cancer-associated decrease in the
CD4/CD8 ratio, as well as a higher proportion of patients
with an inverted CD4/CD8 ratio inside the cancer patient
groups. This phenomenon has been previously described
in very old healthy population [22]. Strindhall and col-
leagues reported a prevalence of 15 % of individuals with
CD4/CD8 ratio below one, in a healthy population beyond
66-year-old [23], similar to the age range used in our study.
CD4 T helper cells take fundamental part in the activation
of CD8 T cells. Thus, the CD4/CD8 inverted ratio may
contribute to explain the reduced capacity of the immune
system in cancer patients to effectively react against patho-
gens [8]. Therefore, the impact of advanced cancer disease
on the immune profile found in our study suggests that
advanced disease may contribute to the immune status, as it
has been recently described [24].
We found a greater proportion of CD28− T cell sub-
population in patients treated with first-line platinum-
based chemotherapy. This is in line with previous results
from Chen and colleagues in a heterogeneous cohort of
four types of cancer patients (respiratory, digestive, repro-
ductive, head and neck) with and without chemotherapy
[8]. It is known that CD28− T cells have a reduced pro-
liferative capacity after activation [25]. Our results sug-
gested that cancer affects B cells and has some impact on
CD4/CD8 ratio, while chemotherapy appears to be more
related to changes in the proportions of CD28− T cell
subpopulations.
Our group has previously reported increasing propor-
tions of CD8 + CD57 + and CD45RA + CD28− T cell
subpopulations with age in healthy subjects [9]. Memory
T cells, particularly those which are specific for regularly
encountered antigens, are in continuous differentiation
toward end-stage as indicated by telomere erosion, replica-
tive senescence and functional exhaustion [16]. The major-
ity of these late-stage differentiated CD28− T cells also
express KLRG-1 [5], CD57 [14] and CD45RA [13] surface
receptors. CD57 expression identifies highly differentiated
13

44
memory T cells with shorter telomere length and reduced
proliferative capacity. CD57 expression is considered
a marker of T cell replicative history, identifying cells
approaching senescence [14].
Likewise, a proportion of these highly differenti-
ated T cells can re-express CD45RA, a marker that has
been previously considered to recognize naive T cells
[13]. It is known that during aging and after CMV infec-
tion there is an increase in highly differentiated CD45RA
re-expressing T cells, within both the CD4 and the CD8
compartments [26]. This remark may reflect the impact
of thymic involution compounded with persistent CMV
infection during aging [27]. Libri et al. [13] have described
that age and CMV serostatus both contribute to the
increase in CD45RA + CD27 T cells in old individuals.
CD45RA + CD27− CD4+ T cells also have significantly
reduced CD28 receptor. The present study showed, for the
first time, that the distribution of CD8 + CD57+ T cells
and CD45RA + CD28− within CD4 and CD8 T cells do
not increase in cancer patients, treated or not with chemo-
therapy, compared with age-matched healthy volunteers.
Hence, supported in the high prevalence of CMV infec-
tion in Cubans [9], our previous and present data suggest,
despite cancer and chemotherapy, that age and probably
persistent CMV infection are mainly responsible for the
increase in CD45RA+ CD28−.
According to our results, those immunosenescence bio-
markers that always change with age show different pat-
terns in the course of cancer disease or with the employ-
ment of first-line platinum-based chemotherapy. Cancer
disease appears to impair B cells and reduce CD4/CD8
ratio. Platinum-based chemotherapy causes high propor-
tions of CD28− T cell subpopulations. Otherwise, changes
in late-stage differentiated CD45RA + CD28− T cells and
the expression of CD57 receptor are not associated with
cancer and chemotherapy.
The variety of factors influencing aging results in a sub-
stantial heterogeneity among people with the same chrono-
logical age, representing a major challenge in daily oncol-
ogy practice [28]. The immune system of cancer patients
can be compromised by the tumor immune suppression
and by previous oncological therapies such as chemother-
apy and radiation. Therefore, a comprehensive evaluation
of patient´s immunocompetence would provide support
for determining whether patients are ready to benefit from
immunotherapy, and perhaps to propose which immuno-
therapy might be suitable for them [29, 30].
Currently, there is a growing interest in the scientific
community for the development and evaluation of biomark-
ers useful for cancer drug development and patient care
[31]. Previous results using CIMAvax-EGF as switch main-
tenance therapy in NSCLC patients have demonstrated that
the magnitude of specific immune response was related to
13
Cancer Immunol Immunother (2016) 65:37–45
the clinical benefit of vaccination and better clinical results
were found in the youngest patients [11, 17]. These results
suggested that a good immune system is required to obtain
better outcomes.
We report here that the proportion of CD8 + CD28−
T cells, CD4 T cells and the CD4/CD8 ratio after first-
line chemotherapy and before the use of CIMAvax-EGF
vaccine impact on the clinical benefit of vaccination in
advanced NSCLC patients. These findings point toward
the potential value of these T cell subpopulation measure-
ments, as predictive biomarkers of CIMAvax-EGF efficacy.
Some examples of targeted drugs and their respective
biomarkers are well known, such as trastuzumab (Her-
ceptin; Roche/Genentech), approved for the treatment
of patients with breast cancer characterized by HER2
amplification and overexpression and crizotinib (Xalkori;
Pfizer), approved by the FDA for the treatment of patients
with NSCLC who carry an ALK gene rearrangement [31].
Recently, Lima and colleagues have proposed a predictive
profile of BCG immunotherapy outcome and a risk score
based on polymorphisms in immune system molecules.
This risk score allows the classification of patients into
risk groups in which BCG treatment is probably successful
[32].
In summary, our results suggest that CD4+ T cells,
CD8+ CD28− T cells and CD4/CD8 ratio are useful tools
as predictive biomarkers of CIMAvax-EGF vaccine effi-
cacy. Further studies need to be done with a larger sample
size to confirm the predictive value of those immunosenes-
cence biomarkers in NSCLC patients treated with CIMA-
vax-EGF vaccine.
Acknowledgments  We thank all participating patients and their
families, as well as staffs of all the institutions involved in this study.
We thank Joel de Leon for critically reviewing the manuscript. The
authors thank Professor Lila Castellanos, Romy Acosta and Mariana
Mansur for the language corrections of the document. Financial sup-
port for the study was provided by Center of Molecular Immunology.
Compliance with ethical standards 
Conflict of interest  The authors declare that they have no conflict
of interest.
References
1. Fulop T, Larbi A, Kotb R, Pawelec G (2013) Immunology of
aging and cancer development. Interdiscip Top Gerontol 38:38–
48. doi:10.1159/000343599
2. Fulop T, Larbi A, Kotb R, de Angelis F, Pawelec G (2011) Aging,
immunity, and cancer. Discov Med 11:537–550
3. Lang PO, Govind S, Aspinall R (2013) Reversing T cell
immunosenescence: why, who, and how. Age 35:609–620.
doi:10.1007/s11357-012-9393-y
4. Pawelec G (2014) Immunosenenescence: role of cytomegalovi-
rus. Exp Gerontol 54:1–5. doi:10.1016/j.exger.2013.11.010
Cancer Immunol Immunother (2016) 65:37–45 45
5. Pawelec G (2012) Hallmarks of human “immunosenes-
cence”: adaptation or dysregulation? Immun Ageing 9:15.
doi:10.1186/1742-4933-9-15
6. Falci C, Gianesin K, Sergi G, Giunco S, De Ronch I, Valpione S
et al (2013) Immune senescence and cancer in elderly patients:
results from an exploratory study. Exp Gerontol 48:1436–1442.
doi:10.1016/j.exger.2013.09.011
7. Mazzola P, Radhi S, Mirandola L, Annoni G, Jenkins M, Cobos
E, Chiriva-Internati M (2012) Aging, cancer, and cancer vac-
cines. Immun Ageing 9:4. doi:10.1186/1742-4933-9-4
8. Chen IH, Lai YL, Wu CL, Chang YF, Chu CC, Tsai IF, Sun FJ,
Lu YT (2010) Immune impairment in patients with terminal can-
cers: influence of cancer treatments and cytomegalovirus infec-
tion. Cancer Immunol Immunother 59:323–334. doi:10.1007/
s00262-009-0753-0
9. Verdecia BG, Hernández DS, Lorenzo-Luaces P, Alvarez TDJB,
Rupalé IL, Herrera ZM, Ramos TC, Dávila AL (2013) Immu-
nosenescence and gender: a study in healthy Cubans. Immun
Ageing 10(1):16. doi:10.1186/1742-4933-10-16
10. Gonzalez G, Diaz-Miqueli A, Crombet T, Raez LE, Lage A
(2013) Current algorithm for treatment of advanced NSCLC
patients: How to include active immunotherapy? J Cancer Ther
4(8A):59–75. doi:10.4236/jct.2013.48A010
11. Neninger Vinageras E, de la Torre A, Osorio Rodríguez M,
Catalá Ferrer M, Bravo I, Mendoza del Pino M, Abreu Abreu
D, Acosta Brooks S, Rives R, del Castillo Carrillo C, González
Dueñas M, Viada C, García Verdecia B, Crombet Ramos T,
González Marinello G, Lage Dávila A (2008) Phase II rand-
omized controlled trial of an epidermal growth factor vaccine in
advanced non-small-cell lung cancer. J Clin Oncol 26(9):1452–
1458. doi:10.1200/JCO.2007.11.5980
12. Fulop T, Larbi A, Witkowski JM, Kotb R, Hirokawa K, Pawelec
G (2013) Immunosenescence and cancer. Crit Rev Oncog
18:489–513
13. Libri V, Azevedo RI, Jackson SE, Di Mitri D, Lachmann R,
Fuhrmann S, Vukmanovic-Stejic M, Yong K, Battistini L,
Kern F, Soares MV, Akbar AN (2011) Cytomegalovirus infec-
tion induces the accumulation of short-lived, multifunctional
CD4  + CD45RA + CD27+ T cells: the potential involvement
of interleukin-7 in this process. Immunology 132:326–339.
doi:10.1111/j.1365-2567.2010.03386.x
14. Appay V, Sauce D, Prelog M (2010) The role of the thymus in
immunosenescence: lessons from the study of thymectomized
individuals. Aging 2:78–81
15. Appay V, van Lier RA, Sallusto F, Roederer M (2008) Phenotype
and function of human T lymphocyte subsets: consensus and
issues. Cytometry A 73:975–983. doi:10.1002/cyto.a.20643
16. Larbi A, Fulop T (2014) From “truly naive” to “exhausted senes-
cent” T cells: when markers predict functionality. Cytom A
85:25–35. doi:10.1002/cyto.a.22351
17. Garcia B, Neninger E, de la Torre A, Leonard I, Martinez R,
Viada C, Gonzalez G, Mazorra Z, Lage A, Crombet T (2008)
Effective inhibition of the epidermal growth factor/epidermal
growth factor receptor binding by anti-epidermal growth factor
antibodies is related to better survival in advanced non-small-
cell lung cancer patients treated with the epidermal growth factor
cancer vaccine. Clin Cancer Res 14:840–846. doi:10.1158/1078-
0432.CCR-07-1050
18. Fulop T, Kotb R, Fortin CF, Pawelec G, de Angelis F,
Larbi A (2010) Potential role of immunosenescence in
cancer development. Ann NY Acad Sci 1197:158–165.
doi:10.1111/j.1749-6632.2009.05370.x
19. Gravekamp C (2011) The impact of aging on cancer vaccination.
Curr Opin Immunol 23:555–560. doi:10.1016/j.coi.2011.05.003
20. Pritz T, Weinberger B, Grubeck-Loebenstein B (2014) The
aging bone marrow and its impact on immune responses in
old age. Immunol Lett 162(1 Pt B):310–315. doi:10.1016/j.
imlet.2014.06.016
21. Fulop T, Montgomery RR (2014) Editorial overview: immune
senescence: known knowns and unknown unknowns. Curr Opin
Immunol 29:7–9. doi:10.1016/j.coi.2014.06.005
22. Olsson J, Wikby A, Johansson B, Lofgren S, Nilsson BO, Fer-
guson FG (2000) Age-related change in peripheral blood T-lym-
phocyte subpopulations and cytomegalovirus infection in the
very old: the Swedish longitudinal OCTO immune study. Mech
Ageing Dev 121:187–201
23. Strindhall J, Skog M, Ernerudh J, Bengner M, Lofgren S, Matus-
sek A, Nilsson BO, Wikby A (2013) The inverted CD4/CD8
ratio and associated parameters in 66-year-old individuals: the
Swedish HEXA immune study. Age 35:985–991. doi:10.1007/
s11357-012-9400-3
24. Whiteside TL (2013) Immune modulation of T-cell and NK (nat-
ural killer) cell activities by TEXs (tumour-derived exosomes).
Biochem Soc Trans 41:245–251. doi:10.1042/BST20120265
25. Akbar AN, Henson SM (2011) Are senescence and exhaustion
intertwined or unrelated processes that compromise immunity?
Nat Rev Immunol 11:289–295. doi:10.1038/nri2959
26. Derhovanessian E, Maier AB, Hahnel K, Beck R, de Craen AJ,
Slagboom EP, Westendorp RG, Pawelec G (2011) Infection
with cytomegalovirus but not herpes simplex virus induces the
accumulation of late-differentiated CD4+ and CD8+ T-cells in
humans. J Gen Virol 92:2746–2756. doi:10.1099/vir.0.036004-0
27. McElhaney JE, Effros RB (2009) Immunosenescence: what does
it mean to health outcomes in older adults? Curr Opin Immunol
21:418–424. doi:10.1016/j.coi.2009.05.023
28. Pallis AG, Hatse S, Brouwers B, Pawelec G, Falandry C, Wed-
ding U, Lago LD, Repetto L, Ring A, Wildiers H (2014) Evaluat-
ing the physiological reserves of older patients with cancer: the
value of potential biomarkers of aging? J Geriatr Oncol 5:204–
218. doi:10.1016/j.jgo.2013.09.001
29. Chang S, Kohrt H, Maecker HT (2014) Monitoring the immune
competence of cancer patients to predict outcome. Cancer Immu-
nol Immunother 63:713–719. doi:10.1007/s00262-014-1521-3
30. Gustafson MP, Lin Y, LaPlant B, Liwski CJ, Maas ML,
League SC, Bauer PR, Abraham RS, Tollefson MK, Kwon ED,
Gastineau DA, Dietz AB (2013) Immune monitoring using the
predictive power of immune profiles. J Immunother Cancer 1:7.
doi:10.1186/2051-1426-1-7
31. Kelloff GJ, Sigman CC (2012) Cancer biomarkers: selecting the
right drug for the right patient. Nat Rev Drug Discov 11:201–
214. doi:10.1038/nrd3651
32. Lima L, Oliveira D, Ferreira JA, Tavares A, Cruz R, Medeiros
R, Santos L (2015) The role of functional polymorphisms in
immune response genes as biomarkers of bacille Calmette-Gué-
rin (BCG) immunotherapy outcome in bladder cancer: establish-
ment of a predictive profile in a Southern Europe population.
BJU Int 116(5):753–63. doi:10.1111/bju.12844
13