American Journal of Otolaryngology–Head and Neck Medicine and Surgery 27 (2006) 76 – 80

www.elsevier.com/locate/amjoto

Original contributions
Decreased hearing after combined modality therapy for
head and neck cancer
Susan E. Pearson, MD, Abby C. Meyer, BS,
George L. Adams, MD, Frank G. Ondrey, MD, PhDT
Department of Otolaryngology, University of Minnesota, SE, Box 396, Minneapolis, MN, USA
Received 4 March 2005

Abstract Purpose: Combined platinum-based chemoradiation therapy is frequently being used as therapy for
head and neck cancer at multiple sites. These therapies are individually ototoxic, but little has been
reported on their combined toxicity.
Materials and methods: A retrospective investigation of 37 patients known to have undergone
therapy with both agents, in combination, for head and neck malignancy was performed. Sixty
percent of the patients had complaints of hearing loss subjectively. Reliable pretreatment and
posttreatment audiograms were obtained on 15 of these patients. Audiograms were analyzed for
sensorineural changes at 0.5, 1, 2, 4, and 8 kHz.
Results: By paired t test analysis, there were significant changes in the patients with pretreatment
and posttreatment audiograms at all frequencies. More than 50% of the patients had a change of
10 dB or greater in their pure-tone average. More than 85% of the patients experienced changes in
their hearing at 4 and 8 kHz.
Conclusions: We conclude that patients undergoing combined modality therapy for head and neck
cancer experience hearing loss. We recommend that hearing assessment, including pretreatment and
posttreatment audiometry, be performed in all patients undergoing combined platinum-based
chemotherapy and radiation for the treatment of head and neck cancer.
D 2006 Elsevier Inc. All rights reserved.

1. Introduction 2 decades. However, CP has well-described ototoxic effects

that have been found to be dose related [5]. This ototoxicity
Recently, platinum-based chemotherapy concurrent with
usually is an irreversible high-frequency sensorineural loss.
radiation therapy (XRT) has become widely used in the
In addition, histological studies show the outer hair cells to
treatment of head and neck malignancy. This is supported
be most susceptible to the ototoxic effects of CP [6]. In
by multiple studies reporting an improved survival for
several human studies, there has been a 20% to 90%
concurrent therapy compared with radiotherapy alone [1- 4].
incidence of hearing loss associated with CP-based regi-
An optimal dose and schedule of these modalities have not
mens [7]. The high variability of this side effect observed in
been agreed upon for head and neck cancer treatment, but
these patients is most likely due to the fact that patients were
parameters have been devised based on toxicities, including
treated with different doses under a range of schedules for a
nephrotoxicity, mucositis, and death secondary to the
variable number of treatment courses. Other factors includ-
intensity of the individual therapeutic regimens.
ing patient age, preexisting hearing loss, interaction with
cis-Platinum (CP) is a chemotherapeutic agent that has
other drugs, and even nutritional status have been suggested
been used in the clinical treatment of cancer for more than
as comorbidity for the clinically observed ototoxicity [8].
T Corresponding author. Department of Otolaryngology, University of
XRT has been commonly used in combination with
Minnesota, SE, Box 396, Minneapolis, MN 55425, USA. Tel.: +1 612 625 surgery for the treatment of head and neck malignancy since
3200; fax: +1 612 625 2101. the 1960s. It has also been shown to be toxic to the middle
0196-0709/$ – see front matter D 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjoto.2005.07.008
 S.E. Pearson et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 27 (2006) 76 – 80 77

Table 1
Overview of patients and platinum and radiation doses
Patient Platinum dose (mg/m2) XRT course XRT dose (cGy) Tumor stage/site Cochlear radiation dose (cGy)
1 75 Split 6186 T3/hypopharynx b 3600
2 100 Continuous 7000 T4/hypopharynx b 3600
3 100 Split 7200 T2/nasopharynx b 7400
4 100 Continuous 6800 T4/supraglottis b 500
5 100 Continuous 7000 T3/base of tongue b 2000
6 100 Split 7000 T4/supraglottis b 500
7 100 Continuous 6000 T2/base of tongue b 2000
8 100 Continuous 7020 T4/maxillary sinus b 7400
9 100 Continuous 7000 T4/supraglottis b 500
10 100 Split 7000 T4/base of tongue, oral pharynx b 2000
11 100 Continuous 7000 T4/pyriform sinus b 3600
12 100 Continuous 7000 T3/tongue b 500
13 100 Continuous 7000 T4/pyriform sinus b 3600
14 100 Continuous 7000 T4/oral cavity b 3200
15 100 Continuous 7000 T4/maxillary sinus b 7400
Patients undergoing split-course radiation had a 1-week treatment break after 3 weeks, and then their second course of CP was simultaneously given with
radiation. Three courses were delivered.

and inner ear both experimentally and in patients undergo- 2. Materials and methods
ing radiation treatment where the fields include the temporal
2.1. Patient information
bone. Radiation otitis, mastoiditis, and osteoradionecrosis in
addition to primary damage to the cochlea have been some In a database review, 37 patients who had completed
of the toxicities described [9]. During radiation treatment treatment with platinum-based chemotherapy in combina-
planning for several head and neck sites, the cochlea and tion with radiation were identified and had completed
temporal bone are included at the margin or within the treatment. All data collection was within the context of
treatment field. As a result, sensorineural hearing loss institutional review board stipulations of the University of
(SNHL) is a known complication of radiotherapy for head Minnesota. Of these patients, 24 had platinum-based
and neck cancer. A recent systematic review of the literature chemotherapy and a complaint of hearing loss during or
found that at least 1 of every 3 patients receiving XRT to the after treatment. Fifteen patients had pretreatment and
nasopharynx or parotid gland develops hearing loss of at posttreatment audiometry at least 1 month after treatment
least 10 dB in the 4-kHz frequency. Fortunately, this loss is completion. Many of the patients were on an unresectable
less pronounced in the lower frequencies [10]. This is in protocol for advanced disease and subsequently died; thus,
agreement with the knowledge that the basal turn of the long-term follow-up audiometry was not performed. The
cochlea, which represents the higher frequencies, is more patients ranged in age from 43 to 85 years with an average
vulnerable to the harmful effects of noise, ototoxic drugs, age of 64.6 F 10.2 years. Thirty-two of the patients were
chemotherapy, and irradiation. Because the outer hair cells men. Physical examinations were conducted as part of their
of the basal turn of the cochlea are arranged in 3 rows routine follow-up examinations for head and neck cancer.
instead of the 4 rows seen in the apical turn, higher Patients who complained of hearing loss or who had middle
frequency sounds are represented by fewer cells. Thus, ear fluid on clinical examination were audiometrically
exposure to ototoxins results in fewer surviving cells in the tested. Patients with fluid and a hearing complaint under-
basal turn than in the apical turn, which is seen as increased went tympanocentesis. Doses of platinum and radiation as
hearing losses at higher frequencies [11]. These losses well as their schedules are shown in Table 1. Expected
would be expected to influence the general hearing of a maximal radiation doses to the ipsilateral cochlea are listed
patient because normal hearing extends to 20 000 Hz for under b Cochlear radiation dose Q and are derived from a
younger patients and to 10 000 Hz for patients at the age of recent analysis of computerized treatment plans for radiation
50 years [12]. by anatomic site [14].
It is accepted that CP interacts with radiation to enhance
2.2. Audiometric testing
tumor cell killing in both aerated and hypoxic cells by
several mechanisms, including the depletion of endogenous Testing was conducted in a double-walled sound booth
radioprotective agents [13]. However, there are few pub- using a Virtual model 320 clinical audiometer (Virtual
lished studies on the combined effects on hearing of CP Corp, Portland, Ore). The audiometric test battery consisted
chemotherapy and XRT. The current retrospective analysis of air and bone conduction measures, speech, and imped-
of our combined modality patients will attempt to describe ance audiometry. Speech was screened with a 25-word list
the incidence and severity of SNHL in patients undergoing (NU-6). During clinical follow-up at the time of treatment
CP/XRT for head and neck cancer. and thereafter, careful questioning about changes in hearing,
78 S.E. Pearson et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 27 (2006) 76 – 80

3. Results
3.1. Posttreatment hearing of all patients
Of the 37 patients who underwent treatment with
platinum and radiation, 24 had complaints of hearing loss.
This group had at least 10-dB decreases in hearing at all
frequencies compared with age- and sex-controlled subjects
from several studies on age-related presbycusis [15]. Fifteen
of these 24 patients with subjective hearing loss had
pretreatment and posttreatment audiograms performed.
The remainder of the analysis presented here will focus on
the findings of the 15 patients with pretreatment and
posttreatment audiograms.
Posttreatment audiometry was performed at least 1 month
after the completion of therapy. Fourteen of the 15 in this
group were men. Audiograms were analyzed for changes in
PTAs in their hearing at 4 or 8 kHz. The number of patients
with frequency-specific hearing loss is shown in Fig. 1.
Typically, platinum is not expected to have significant
effects at the lower frequencies, but 8 of 15 patients did
experience a loss of 10 dB or greater in their PTA. The
average loss in PTA for this group was 10.0 F 12.5 dB.
Paired t test analysis for this increase in PTA for the group
was also significant ( P b .001). These results are depicted
in Fig. 2. Of note, patients 3, 8, and 15 from Table 1 were
the only patients who received greater than 4000 cGy to
either of their cochleae, a level that might cause toxicity to
neural structures. In a subset analysis of these 3 patients,
however, only 1 of the 3 experienced a significant increase
Fig. 1. Patients with hearing loss of 10 dB or greater after combined in their PTA. At 4 kHz, all patients had a loss of 10 dB or
radiation and platinum. Eight of 15 patients experienced loss of their PTA. greater compared with pretreatment audiometry. The aver-
Fifteen of 15 and 13 of 15 patients had losses at 4 and 8 kHz, respectively. age hearing loss in the entire group was 26.5 F 22.3 dB
compared with pretreatment values. This average hearing
loss for the group at 4 kHz was highly significant by paired
tinnitus, and vertigo was conducted in these patients. The t test analysis ( P b 1  10 6). This is also shown in Fig. 2.
24 patients with hearing complaints had audiometry Similarly, at 8 kHz, 13 of 15 patients experienced a loss of
performed at the time of the complaint and at later patient 10 dB or greater as well. For the entire group, the average
visits. The hearing status of some patients dictated further
clinical testing after the completion of treatment. Nine of
the 24 patients with complaints did not undergo pretreat-
ment audiometry.

2.3. Data analysis

Audiograms were analyzed for fluctuations in hearing at
all standard (0.25, 0.5, 1, 2, 3, 4, 6, and 8 kHz) frequencies.
Audiograms were also examined for increases in the pure-
tone average (PTA; 0.5, 1, and 2 kHz). Changes in speech
discrimination were also analyzed. None of the audiometric
data presented here reflect threshold shifts secondary to
middle ear fluid. Any increases of 10 dB or greater in the
PTA in their hearing at 4 or 8 kHz were considered
significant. Decreases in speech recognition of 20% or
greater were considered significant. Because the speech
reception threshold data were not consistently collected in Fig. 2. Average hearing loss by frequency in 15 patients treated with
each case, it was not used in the data analysis. Paired t test platinum chemotherapy and XRT. All changes were significant at P b .01
analysis was performed on these 15 patients as well. or less by paired t test analysis.
 S.E. Pearson et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 27 (2006) 76 – 80 79

Table 2 SNHL [16,17]. In our study, the platinum dose received by

Incidence of frequency-specific hearing loss by severity in 15 patients with the patients was less than that expected to cause ototoxicity;
pretreatment and posttreatment audiometry after treatment with CP and
radiation
however, half the patients had PTA loss, all the patients had
loss at 4 kHz, and 13 of 15 patients had loss at 8 kHz.
Frequency Loss of 20 dB or greater Loss of 30 dB or greater
One recent study [18] looked at the rates of SNHL in
PTA 26% (4/15) 7% (1/15)
32 patients, all with nasopharyngeal cancer, who received
4 kHz 73% (11/15) 53% (8/15)
8 kHz 67% (10/15) 53% (8/15) 1 cycle of neoadjuvant chemotherapy with 5-fluorouracil
(1000 mg/m2 for 4 days) and cisplatin (20 mg/m2 for 4 days)
followed by 2 to 3 cycles of cisplatin (20 mg/m2 for 4 days)
hearing loss was 23.6 F 21 dB. This loss was also highly every 3 weeks concurrently during XRT (total dose, 70 Gy).
significant by paired t test analysis ( P b 1  10 5). The During the follow-up, 16.7% of ears showed at least 15-dB
losses in PTA were assessed as well. No patients in the study
loss in the PTA, and 43.8% showed at least 15-dB loss at
experienced a significant loss in their word discrimination
4 kHz. The authors concluded, based on historical data on
(N 20% decrease). However, several patients did require an
SNHL after radiotherapy alone, that the incidence and
increase in the amplitude at which the word list was
features of SNHL after combined chemoradiation therapy
presented to achieve a similar score as before treatment. The
with cisplatin were similar to SNHL after XRT alone in the
number of patients who had greater than 20- and 30-dB
treatment of nasopharyngeal carcinoma.
losses for a given frequency was also examined. Four of
In our study, 65% (24/37) of the total original patients
15 patients had an increase in their PTA of greater than
had subjective hearing loss that prompted audiometric
20 dB, and 1 patient had a greater than 30-dB increase in the
PTA. Most of the patients also had greater than 30-dB losses testing. Fifteen of these patients had pretreatment and
at 4 and 8 kHz (8/15 patients at both frequencies). These posttreatment audiometric analysis and were available for
data are depicted in Table 2. complete analysis in this investigation. At least 6 (16%) of
37 patients experienced losses of 15 dB or greater in the
3.2. Other findings PTA. This is similar to the 16.7% reported in the
aforementioned study on nasopharyngeal chemoradiation
In this series, only 1 patient developed a perforation that
persisted after treatment. Also of note, 6 of the 37 patients [18]. However, in the study by Oh et al [18], all patients
developed serous otitis requiring tympanocentesis. were treated for cancer of the nasopharynx, and thus, it is
likely that 100% of them received more than 7000 cGy to at
least 1 of their cochleae, based on previously published data
4. Discussion [14]. In our study only 3 of 15 patients had levels of
radiation to the cochlea greater than 4000 cGy, a dose that
The ototoxicity of platinum and radiation in combination
for the treatment of head and neck neoplasia was analyzed may cause toxicity to neural structures, and yet, more than
in this retrospective review. Because the use of this half of the patients had PTA loss, all the patients had loss at
combined therapy has increased in recent years, their 4 kHz, and 13 of 15 patients had loss at 8 kHz. Thus,
combined ototoxicity requires clinical study. radiation alone cannot explain the hearing loss experienced
In the present study, all of the patients were treated on by the patients in our study.
head and neck cancer protocols where platinum was given at The amount of hearing loss our patients experienced was
75 to 100 mg/m2 at defined intervals before or during the much higher than expected based on the amount of CP they
radiation treatment. All patients received 3 cycles at this received and the amount of radiation received by their
dose, except for 1 patient. Prior studies indicate that the cochleae. Studies have shown that adding low-dose cisplatin
toxicity of platinum increases at total doses starting at to XRT does not increase the risk of SNHL over that with
approximately 400 mg [16]. These analyses do not account XRT alone [18,19]. However, these studies looked only at
for patient weight and body surface area; thus, calculated patients treated for squamous cell carcinoma of the
total dose does not reflect plasma platinum concentrations. nasopharynx, where the radiation dose to the cochlea is
In no study were plasma levels of platinum followed. The already high enough to cause SNHL, with or without
interval between treatment courses, the numbers of cycles concurrent cisplatin. This was not the case in our study, in
per patient, dose per cycle, and bolus versus continuous which only 1 patient had disease localized to the nasophar-
administration are among factors that govern the pharma- ynx. The higher than expected amount of hearing loss our
cokinetics of platinum and may predict the eventual patients experienced suggests the likelihood that combined
ototoxicity. In most chemoradiation therapy protocols for CP and XRT may produce more ototoxicity than XRT alone
head and neck cancer, the dose of platinum ranges from 50 to when the cochlea has not already been extensively damaged
100 mg/m2 and is given every 2 weeks for 3 cycles. It has by high amounts of radiation.
previously been published that 150 mg/m2 dosing or total Another important finding in this study is that the
doses of 400 mg or greater can cause loss of cochlear sensorineural loss experienced by these patients is likely
function; thus, these protocols should produce low levels of at the level of the cochlea exclusively. Patients did not have
80 S.E. Pearson et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 27 (2006) 76 – 80

changes in their discrimination scores that would be [3] Al-Sarraf M, LeBlanc M, Giri PG, et al. Chemoradiotherapy versus
suggestive of retrocochlear effects of the treatment regimen. radiotherapy in patients with advanced nasopharyngeal cancer:
phase III randomized intergroup study 0099. J Clin Oncol 1998;
In summary, the present study indicates that a similar 16(4):1310 - 7.
percentage of our patients experienced SNHL compared with [4] Glicksman AS, Slotman G, Doolittle III C, et al. Concurrent cis-
patients reported by Oh et al [18] being treated with platinum and radiation with or without surgery for advanced head and
concurrent cisplatin and XRT for nasopharyngeal carcinoma. neck cancer. Int J Radiat Oncol Biol Phys 1994;30(5):1043 - 50.
These results were similar despite our patients having disease [5] Skinner R, Pearson AD, Amineddine HA, et al. Ototoxicity of
cisplatinum in children and adolescents. Br J Cancer 1990;61:
in multiple anatomic sites besides the nasopharynx. Further- 927 - 31.
more, because patients without pretreatment audiograms who [6] Sie KC, Norton SJ. Changes in otoacoustic emissions and auditory
had a subjective hearing loss were excluded from analysis, it brain stem response after cis-platinum exposure in gerbils. Otolar-
is likely that the actual number of patients experiencing yngol Head Neck Surg 1997;116(6 Pt 1):585 - 92.
hearing loss in the present account is underreported. [7] Morose M, Blair RJ. A review of cis-platinum ototoxicity. Otolaryn-
gology 1983;12:365 - 9.
The potential for hearing loss in these regimens must be [8] Lauterman J, Song B, McLaren J, et al. Diet is a risk factor in cis-
recognized, and more research into ways to minimize platinum ototoxicity. Hear Res 1995;88:47 - 53.
cochlear radiation must be undertaken. Pretreatment CT [9] Borsanyi S. The effect of radiation on the ear. South Med J 1962;
planning, XRT port design, and temporal bone shielding are 55:740 - 3.
possible ways to reduce exposure of the middle and inner [10] Raaijmakers E, Engilen AM. Is sensorineural hearing loss a possible
side effect of nasopharyngeal and parotid irradiation? A systematic
ear while still achieving adequate coverage of the tumor and review of the literature. Radiother Oncol 2002;65:1 - 7.
its lymphatic drainage. However, these techniques are not [11] Ho WK, Wei WI, Kwong DLW, et al. Long-term sensorineural
routinely used. Furthermore, it is possible that those hearing deficit following radiotherapy in patients suffering from
experiencing significant ototoxicity with the CP chemo- nasopharyngeal carcinoma: a prospective study. Head Neck 1999;
radiation regimen could receive carboplatinum as a less 21:547 - 53.
[12] Honore HB, Bentzen SM, Moller K, et al. Sensori-neural hearing loss
ototoxic agent for subsequent cycles during XRT. after radiotherapy for nasopharyngeal carcinoma: individualized risk
SNHL is frequently seen in patients receiving combined estimation. Radiother Oncol 2002;65:9 - 16.
CP and XRT for the treatment of nasopharyngeal carcinoma, [13] Douple E. Interactions between platinum coordination complexes and
but little is reported in the literature about the regimen’s radiation. Antitumor drug radiation interactions. Boca Raton (Fla)7
ototoxicity in treating other anatomic sites within the head CRC Press; 1990. p. 171 - 90.
[14] Ondrey FG, Greig JR, Herscher L. Radiation dose to otologic
and neck. As a result, it would be reasonable to establish structures during head and neck cancer radiation therapy. Laryngo-
hearing screening as part of all protocols involving platinum- scope 2000;110(2 Pt 1):217 - 21.
based chemoradiation therapy of the head and neck. It is [15] Spoor A. Presbycusis values in relation to noise induced hearing loss.
important to recognize toxicities that may impact the quality Audiol Int 1967;9:48 - 57.
of life in the cohort who may be long-term survivors. [16] Schaefer S, Post J, Close L, et al. Ototoxicity of low and moderate
dose cisplatin. Cancer 1985;56:1934 - 9.
[17] Kopelman J, Budnick AS, Sessions RB, et al. Ototoxicity of high-dose
References cisplatin by bolus administration in patients with advanced cancers
and normal hearing. Laryngoscope 1988;98:858 - 64.
[1] Browman GP, Hodson D, Mackenzie RJ, et al. Choosing a [18] Oh YT, Kim CH, Choi JH, et al. Sensory neural hearing loss after
concomitant chemotherapy and radiotherapy regimen for squamous concurrent cisplatin and radiation therapy for nasopharyngeal carci-
cell head and neck cancer: a systematic review of the published noma. Radiother Oncol 2004;72:79 - 82.
literature with subgroup analysis. Head Neck 2001;23:579 - 89. [19] Kwong DL, Wei WI, Sham JS, et al. Sensorineural hearing loss in
[2] Brizel DM, Albers ME, Fisher SR, et al. Hyperfractionated irradiation patients treated for nasopharyngeal carcinoma: a prospective study of
with or without concurrent chemotherapy for locally advanced head the effect of radiation and cisplatin treatment. Int J Radiat Oncol Biol
and neck cancer. N Engl J Med 1998;338:1798 - 804. Phys 1996;36(2):281 - 9.