Jpn J Clin Oncol 2008;38(4)244249

doi:10.1093/jjco/hyn022

Original Articles

Capecitabine Monotherapy for Recurrent and Metastatic

Nasopharyngeal Cancer
Daniel Chua1, William I. Wei2, Jonathan S.T. Sham1 and Gordon K.H. Au1
1

Department of Clinical Oncology, The University of Hong Kong, Queen Mary Hospital, Hong Kong SAR and
Department of Surgery, The University of Hong Kong, Queen Mary Hospital, Hong Kong SAR, China

Received December 27, 2007; accepted February 22, 2008

Key words: nasopharyngeal carcinoma capecitabine hand-foot syndrome palliative

chemotherapy

INTRODUCTION
Nasopharyngeal carcinoma (NPC) is highly responsive to
chemotherapy and effective palliation of symptoms, and prolongation of survival can often be achieved after chemotherapy for recurrent and metastatic disease with occasional
observation of long-term survivors after treatment (1,2).
Platinum-based chemotherapy regimens are among the most
effective regimens for NPC and they are often used as rstline chemotherapy for recurrent or metastatic disease. For
For reprints and all correspondence: Daniel Chua, Department of Clinical
Oncology, PB-115, Queen Mary Hospital, Pokfulam, Hong Kong, China.
E-mail: [email protected]

those with disease recurrence or progression after prior

platinum-based chemotherapy, there is no standard secondline chemotherapy although several regimens have been
reported to achieve a response rate of 29 56% in platinumpretreated patients (3 5). In patients who received multiple
lines of chemotherapy, toxicity became a major issue and
the identication of active regimen with a favorable toxicity
prole is important.
Capecitabine is a novel oral tumor-activated and -elective
uoropyramidine carbamate. The drug exploits the signicantly higher thymidine phosphorylase activity of malignant
cells as compared with healthy cells for preferential conversion of capecitabine to 5-uorouracil in malignant cells.

# The Author (2008). Published by Oxford University Press. All rights reserved.

Downloaded from http://jjco.oxfordjournals.org/ at Kainan University on March 24, 2015

Background: Capecitabine monotherapy had activity in recurrent/metastatic nasopharyngeal

carcinoma (NPC) as demonstrated previously in a small pilot study. We conducted a retrospective review of patients who received capecitabine for recurrent and metastatic NPC to
further evaluate its clinical benefits.
Methods: Forty-nine patients with recurrent and metastatic NPC received capecitabine at a
dose of 1 1.25 G/m2 twice daily for 14 days in 3-week cycles. Disease sites were locoregional in 29%, distant in 45% and locoregional plus distant in 26%. All except one had prior platinum-based chemotherapy for relapse or as adjunctive treatment. Median follow-up was 10
months (range: 3 41).
Results: Treatment was generally well tolerated. Hand-foot syndrome was common and
occurred in 86% (25% Grade 3). Grade 3 hematological toxicity occurred in 6%. Partial
response rate was 31% (95% CI: 18%, 44%) and complete response rate was 6% (95% CI:
0%, 13%), for an overall response rate of 37% (95% CI: 23%, 50%). Median time-to-progression was 5 months and median survival was 14 months. One- and two-year survival
rates were 54 and 26%, respectively. Significantly better survival was observed in patients
treated for locoregional recurrence and those with severe hand-foot syndrome.
Conclusions: Capecitabine has single agent activity in NPC and severe hand-foot syndrome
predicts favorable outcome. Based on our experience, capecitabine monotherapy should be
considered in patients with recurrent/metastatic NPC.

Jpn J Clin Oncol 2008;38(4)

The selective tumor-activation of capecitabine allows continuous tumor exposure to 5-uorouracil while reducing side
effects by reducing exposure of normal tissues to the drug.
The convenience of oral administration provides an added
advantage. We have previously conducted a Phase II trial
using capecitabine in 17 patients with recurrent or metastatic
NPC pretreated with platinum-based chemotherapy, and we
observed a response rate of 23.5% with a median
time-to-progression of 4.9 months (6). On the basis of the
results of this trial, capecitabine monotherapy was used as
palliative chemotherapy for advanced NPC in our institution.
To further explore the efcacy of capecitabine and to identify possible predictive factors of response, we conducted a
retrospective review of the outcome of our patients who
received capecitabine for recurrent or metastatic NPC.

Between January 2004 and December 2006, 49 patients

received capecitabine monotherapy for recurrent and/or
metastatic NPC at Queen Mary Hospital. The ratio of male
to female was 2.3:1, and the median age was 46. All patients
except one, who refused intravenous chemotherapy, had previously received platinum-based chemotherapy as adjunctive
treatment or palliative treatment for relapse. Disease sites
were local and/or regional in 29%, distant only in 45% and
locoregional plus distant in 26%. About 22% of patients
received capecitabine as rst-line chemotherapy for relapse,
45% as second-line and 33% as third-line treatment.
Table 1 summarizes the patient and disease characteristics.
Capecitabine was administered orally at a dose of 1
1.25 G/m 2 twice daily in 3-week cycles consisting of 2
weeks of treatment followed by rest period of 1 week.
Thirty-seven patients (75.5%) received capecitabine at a
standard dose of 1.25 G/m2 twice daily. Since a high proportion of patients treated with the standard dose developed
severe hand-foot syndrome requiring dose modication, the
starting dose of capecitabine was subsequently reduced to
1 G/m2 twice daily and 12 patients were treated using the
reduced dose. In the event of severe hand-foot syndrome,
capecitabine dose was modied as follows: no dose change
for Grade 1 toxicity; for Grade 2 toxicity, no dose change at
rst occurrence of event, 25% dose reduction at second
occurrence and 50% dose reduction at third occurrence; for
Grade 3 toxicity, 25% dose reduction at rst occurrence of
event and 50% dose reduction at second occurrence.
Treatment was interrupted when patient developed Grade 2
3 hand-foot syndrome and resumed after resolution of toxicity to Grade 0 1.
Complete blood picture and serum biochemistry were routinely tested before each cycle. Patients were assessed for
toxicity before each cycle of treatment. Toxicity was graded
according to the Common Terminology Criteria for Adverse
Events version 3.0. Tumor response was assessed using
Response Evaluation Criteria in Solid Tumors by imaging

Table 1. Characteristics of patients who received capecitabine for recurrent

and/or metastatic nasopharyngeal carcinoma
Gender (n)
Male

34 (69%)

Female

15 (31%)

Age (years)
Median

46

Range

292

Karnofsky performance score

Median

75

Range

60100

No. of relapse (n)

First relapse

31 (63%)

Second relapse

15 (31%)

Third relapse

3 (6%)

Relapse pattern (n)

Locoregional

14 (29%)

Distant

22 (45%)

Locoregional and distant

13 (26%)

Previous chemotherapy received (n)

Platinum

48 (98%)

Taxane

10 (20%)

Gemcitabine

11 (22%)

Ifosfamide

14 (29%)

5-uorouracil

38 (78%)

No. of chemotherapy regimens received (n)

One

24 (49%)

Two

21 (43%)

 three

3 (6%)

before treatment, after third and sixth cycles, and thereafter

every 3 4 cycles. Patients who had treatment discontinued
but without disease progression had follow-up imaging at 2
and 4 months after treatment and every 3 4 months thereafter. Treatment was continued until disease progression, presence of unacceptable toxicity, or patient refusal.
All patients were included in efcacy and safety analysis.
The following endpoints were used in the evaluation of efcacy: response rate, progression-free survival (PFS) and
overall survival (OS). PFS and OS probabilities were calculated by the Kaplan Meier method. Time-to-progression
was calculated from the date of rst dose of capecitabine to
the date of documented progression or most recent follow-up
in patients without disease progression. Survival time was
calculated from the date of rst dose of capecitabine to the
date of death or most recent follow-up for surviving patients.
Univariate analysis was performed to study the predictive
value of different variables on PFS and OS, using log rank
test to compare the differences between actuarial curves.

Downloaded from http://jjco.oxfordjournals.org/ at Kainan University on March 24, 2015

PATIENTS AND METHODS

245

246

Capecitabine for nasopharyngeal carcinoma

Response rates among different subgroups were compared

using chi-square test or Fishers exact test as appropriate. A
value of P , 0.05 was considered to be statistical signicant.

RESULTS
TREATMENT OUTCOME

TOXICITY OF TREATMENT
Treatment was generally well tolerated. Table 2 summarizes
the toxicity of capecitabine treatment. Hematological toxicity
was mild with only three patients developing Grade 3 toxicity (6%), including Grade 3 anemia in two patients and
Grade 3 thrombocytopenia in one patient. The most common

Figure 2. Overall survival curve in recurrent and metastatic nasopharyngeal

carcinoma patients treated with capecitabine.

non-hematological toxicity was hand-foot syndrome, which

occurred in 42 patients (86%), with Grade 3 toxicity in 12
patients (25%), despite the use of prophylactic topical emollient creams and pyridoxine at a dose of 100 mg/day.
Patients with hand-foot syndrome were treated with topical
emollient creams /2 antibiotic creams. Thirteen patients
(27%) had dose modication during treatment, all due to
hand-foot syndrome. There was no signicant difference in
the percentage of patients, who developed severe hand-foot
syndrome using a starting dose of 1.25 or 1 G/m2 (58 versus
57%), and the percentage of patients with dose modication
was similar with a starting dose of 1.25 or 1 G/m2 (25 versus
27%). Not all patients with Grade 2 3 hand-foot syndrome
had dose modication due to unwillingness of some patients
to have their treatment dose reduced. Although the incidence
of hand-foot syndrome was high, none of our patients had
treatment discontinued due to this toxicity alone. None of
Table 2. Summary of toxicity of capecitabine treatment in patients with
nasopharyngeal carcinoma
Number of patients
Grade 0
Anemia

Grade 2
5

Grade 3
2

Neutropenia

39

Thrombocytopenia

44

Diarrhea

42

Stomatitis

36

11

14

16

12

35

12

Hand-foot syndrome
Nausea

Figure 1. Progression-free survival curve in recurrent and metastatic nasopharyngeal carcinoma patients treated with capecitabine.

33

Grade 1

Vomiting

42

Fatigue

35

10

Rash

47

Downloaded from http://jjco.oxfordjournals.org/ at Kainan University on March 24, 2015

Patients received 1 10 cycles of capecitabine with a median

of four cycles. Median follow-up time was 10 months
(range: 3 41). Treatment was discontinued in 16 patients
(33%) due to disease progression. In the remaining 33
patients (67%), treatment was discontinued in the absence of
disease progression due to patient decision, either because of
lack of further response or nancial issue since capecitabine
cost was not covered by the hospital budget. Best response
to capecitabine was complete response in three patients (6%;
95% condence interval (CI): 0%, 13%), partial response in
15 patients (31%; 95% CI: 18%, 44%), stable disease in 21
patients (43%; 95% CI: 29%, 57%) and progressive disease
in 10 patients (20%; 95% CI: 9%, 32%). Overall response
(OR: complete partial) rate was 37% (95% CI: 23%,
50%). OR rate according to disease sites was 46% for local
disease, 36% for regional disease and 31% for distant metastases. PFS rate was 37% at 6 months and 9% at 12 months
(Fig. 1), with a median time-to-progression of 5 months. OS
rate was 54% at 1 year and 26% at 2 years (Fig. 2), with a
median survival of 14 months.

Jpn J Clin Oncol 2008;38(4)

the patients developed any Grade 4 toxicity and there were

no treatment-related deaths.

Table 3. Summary of treatment outcome according to disease and treatment

factors
Overall
response
rate

Progression-free
survival rate
(6-month)

Overall
survival
rate
(1-year)

Male (%)

41

38

52

Female (%)

27

36

57

ADDITIONAL TREATMENT AFTER CAPECITABINE

Sixteen patients (33%) received further intravenous chemotherapy after discontinuation of capecitabine, usually at
the time when disease progression was documented. The
remaining 33 patients (67%) did not receive further chemotherapy after capecitabine and were put on supportive
care only.

PREDICTIVE FACTORS

OF

TREATMENT OUTCOME

Gender

P value

0.33

Chemotherapy remains an important palliative treatment for

recurrent and/or metastatic NPC. Although no randomized
trial comparing different chemotherapy regimens in NPC has
ever been reported, platinum/5-uorouracil combination was
commonly employed due to the high response rate observed
in chemotherapy-nave patients. Combined chemo-radiotherapy
has now been accepted as standard treatment for locoregionally advanced NPC, and many centers also routinely offer
chemo-radiotherapy for patients with Stage II disease. As a
consequence, most patients with recurrent or metastatic
disease nowadays have already been pretreated with platinum
in the adjunctive setting. For these patients, as well those
who progressed after prior platinum-based chemotherapy,
there is no agreement on standard chemotherapy regimen
that should be used. Taxane- or gemcitabine-based doublet
combination was commonly employed, with reported
response rate of 22 75% in the former (3,7 9) and 36 75%
in the latter (5,10,11), but hematological toxicity was
common and dose-limiting.

0.45

 45 (%)

33

26

52

. 45 (%)

40

48

54

0.63

0.26

0.80

Relapse pattern
Locoregional only (%)

50

62

79

Distant only (%)

32

36

45

Locoregional distant (%)

31

15

40

P value

0.48

0.023

0.014

Capecitabine as
First-line Rx (%)

36

18

53

Second-line Rx (%)

23

34

45

 third-line Rx (%)

56

63

65

P value

0.11

0.040

0.44

Prior 5-uorouracil Rx
Yes (%)

32

41

46

No (%)

55

27

80

P value

0.29

0.33

0.46

Hand-foot syndrome
Grade 0/1 (%)

19

33

32

Grade 2/3 (%)

50

40

71

P value

0.026

0.13

0.002

Capecitabine dose
2 G/m2 daily (%)

25

31

57

41

39

53

2.5 G/m daily (%)

P value

0.49

0.68

0.35

Rx, treatment.

Palliative chemotherapy using triplet or more combination

has also been explored. In one report, combination of cyclophosphamide, doxorubicin, cisplatin, methotrexate and bleomycin was used for chemotherapy-nave NPC patients (12).
The response rate in patients with recurrence and metastatic
disease was 41 and 80%, respectively. The corresponding
median survival was 16 and 14 months. Toxicity was,
however, severe with 41% patients requiring admission for
management of toxicity and treatment-related deaths
occurred in 11%. In another report, combination of paclitaxel, carboplatin and gemcitabine was used in 32 metastatic
NPC patients and yielded a response rate of 78%, but Grade

Downloaded from http://jjco.oxfordjournals.org/ at Kainan University on March 24, 2015

DISCUSSION

0.54

Age

P value

Treatment outcome was analysed according to different

disease and treatment factors and the results were summarized in Table 3. No signicant difference in OR rate was
observed among patients treated for different relapse patterns. Both PFS and OS rates were, however, signicantly
better in those treated for locoregional recurrence only
(Fig. 3). No signicant difference in OR and OS rates was
observed among patients who received capecitabine as rst,
second or greater than or equal to third-line of chemotherapy
for relapse, but PFS rate was lower when capecitabine was
used as rst-line treatment for relapse. No signicant difference in treatment outcome was observed in patients with and
without prior treatment with 5-uorouracil. Patients with
Grade 2 3 hand-foot syndrome had signicantly better OR
and OS rates than those with Grade 0 1 toxicity (Fig. 4).
Treatment outcome in patients treated with the reduced starting dose of capecitabine was not signicantly different from
those treated with the standard dose.

247

248

Capecitabine for nasopharyngeal carcinoma

3 4 neutropenia was common and occurred in 78% of

patients (13).
Single agent chemotherapy other than platinum was generally considered to be of limited benet in NPC. In heavily
pretreated patients, single agent chemotherapy is likely to be
more tolerable and acceptable to patients. Agents, such as
mitoxantrone (14) and irinotecan (15), have been studied but
the response rate was low. Even when the active agent gemcitabine was used alone, the response rate appears to be
lower than that of the combination regimen containing the
same drug (16). A Phase II study using capecitabine alone
as palliative chemotherapy in pretreated NPC patients was
conducted by us previously, and a response rate of 23.5%
was observed. The current report represents our experience
of using capecitabine in routine clinically setting, and the
response and survival rates observed were similar to our previous report.

Figure 4. Comparison of overall survival curves in recurrent and metastatic

nasopharyngeal carcinoma patients with and without severe hand-foot syndrome to capecitabine.

Downloaded from http://jjco.oxfordjournals.org/ at Kainan University on March 24, 2015

Figure 3. Comparison of overall survival curves in recurrent and metastatic

nasopharyngeal carcinoma patients with different relapse pattern.

Most of our patients found capecitabine to be convenient

and tolerable. Hand-foot syndrome, however, was common
and the toxicity had signicant impact on patient compliance, although it was reversible and always improved after
dose interruption and modication. Owing to the relatively
mild toxicity, capecitabine can actually be administered for
prolonged cycles in the absence of disease progression. In
the current series, most patients had treatment discontinued
in the absence of disease progression, and it is uncertain
whether prolonged treatment can further improve the
outcome.
Our observation of improved outcome in patients treated
for locoregional recurrence only was likely due to the different clinical course in this group compared with patients
treated for distant metastases. Similar to the ndings of our
previous Phase II trial, we also observed the association of
response to capecitabine with severe hand-foot syndrome in
the current series. This association was not related to capecitabine dose or treatment duration, as most patients developed
the worst reaction by the end of third cycles before assessment of response. In view of the observed association of
severe hand-foot syndrome and subsequent tumor response,
it is important to reassure patient and not to discontinue
treatment prematurely if severe toxicity occurs early in the
course of treatment.
Table 4 summarizes the results of using different chemotherapy regimens for recurrent and metastatic NPC
reported in the literature. Although one should be careful in
making direct comparison of these results due to different
patient and disease characteristics, the data listed are in
agreement with clinical experience. Capecitabine monotherapy compares favorably with other chemotherapy regimens
in terms of efcacy, even if majority of our patients did not
receive further chemotherapy after capecitabine and almost
all were platinum pretreated. On the other hand, resistant
clones are more likely to emerge early with single agent
than combination chemotherapy, patients with newly diagnosed relapse should benet more from combination chemotherapy as initial treatment. The signicantly poorer PFS
rate, when capecitabine was used as rst-line treatment
observed in our study, suggests that capecitabine should be
reserved for second- or third-line treatment for recurrent/
metastatic NPC, and combination chemotherapy should be
employed instead as rst-line treatment.
In summary, this retrospective study conrmed the ndings of our previous Phase II trial about the single agent
activity of capecitabine in recurrent and metastatic NPC.
Presence of severe hand-foot syndrome predicts good
response and better survival after treatment. Combining
capecitabine with other chemotherapeutic agents may further
improve the efcacy and is worth investigating. Two multicenter Phase II studies testing the combination of capecitabine plus cisplatin and capecitabine plus oxaliplatin in
recurrent and metastatic NPC are currently being conducted
in Asian regions. Based on our experience, capecitabine
monotherapy should be considered in patients with recurrent

Jpn J Clin Oncol 2008;38(4)

249

Table 4. Summary of reported response rate and treatment outcome after chemotherapy using different regimens for recurrent/metastatic NPC
Chemotherapy

Response rate (%)

Median progression-free survival

Median survival

Yeo et al. (3)

Paclitaxel/carboplatin

59

6.0 months

13.9 months

Tan et al. (7)

Paclitaxel/carboplatin

75

7.0 months

12.0 months

McCarthy et al. (9)

Docetaxel/cisplatin

22

8.4 months

not reported

Chua et al. (8)

Docetaxel/cisplatin

63

5.6 months

12.4 months

Ngan et al. (5)

Gemcitabine/cisplatin

73

10.6 months

15 months

Chua et al. (10)

Gemcitabine/cisplatin

75

not reported

not reported

Wang et al. (11)

Gemcitabine/vinorelbine

36

5.6 months

11.9 months

Foo et al. (15)

Gemcitabine (chemotherapynave)

28

3.6 months

7.2 months

Foo et al. (15)

Gemcitabine (pretreated)

48

5.1 months

10.5 months

Dugan et al. (14)

Mitoxantrone

25

2.7 months

13.0 months

Poon et al. (16)

Irinotecan

17

3.9 months

11.4 months

Current report

Capecitabine

37

5 months

14 months

NPC, nasopharyngeal carcinoma.

and/or metastatic NPC after prior treatment with other chemotherapy regimens.
Conict of interest statement
None declared.

References
1. Teo PML, Kwan WH, Lee WY, et al. Prognosticators determining
survival subsequent to distant metastases from nasopharyngeal
carcinoma. Cancer 1996;77:2423 31.
2. Fandi A, Bachouchi M, Azli N, et al. Long-term disease-free survivors
in metastatic undifferentiated carcinoma of the nasopharyngeal type.
J Clin Oncol 2000;18(6):132430.
3. Yeo W, Leung TW, Chan AT, et al. A phase II study combination of
paclitaxel and carboplatin in advanced nasopharyngeal carcinoma. Eur
J Cancer 1998;34(13):202731.
4. Chua DT, Kwong DL, Sham JS, et al. A phase II study of ifosfamide,
5-uorouracil and leucovorin in patients with recurrent nasopharyngeal
carcinoma previously treated with platinum chemotherapy. Eur J
Cancer 2000;36:73641.
5. Ngan RK, Yiu HH, Lau WH, et al. Combination gemcitabine and
cisplatin chemotherapy for metastatic or recurrent nasopharyngeal
carcinoma: report of a phase II study. Ann Oncol 2002;13:1252 8.
6. Chua DT, Sham JS, Au GK. A phase II study of capecitabine in
patients with recurrent and metastatic nasopharyngeal carcinoma
pretreated with platinum-based chemotherapy. Oral Oncol 2003;39:
361 6.

7. Tan EH, Khoo KS, Wee J, et al. Phase II trial of a paclitaxel and
carboplatin combination in Asian patients with metastatic
nasopharyngeal carcinoma. Ann Oncol 1999;10:235 7.
8. Chua DT, Sham JS, Au GK. A phase II study of docetaxel and cisplatin
as rst-line chemotherapy in patients with metastatic nasopharyngeal
carcinoma. Oral Oncol 2005;41:58995.
9. McCarthy JS, Tannock IF, Degendorfer P, et al. A Phase II trial of
docetaxel and cisplatin in patients with recurrent or metastatic
nasopharyngeal carcinoma. Oral Oncol 2002;38:68690.
10. Chua DT, Sham JS, Au GH. Induction chemotherapy with cisplatin and
gemcitabine followed by reirradiation for locally recurrent
nasopharyngeal carcinoma. Am J Clin Oncol 2005;28:464 71.
11. Wang CC, Chang JY, Liu TW, et al. Phase II study of gemcitabine plus
vinorelbine in the treatment of cisplatin-resistant nasopharyngeal
carcinoma. Head Neck 2006;28:7480.
12. Siu LL, Czaykowshi PM, Tannock IF. Phase I/II study of the
CAPABLE regimen for patients with poorly differentiated carcinoma of
the nasopharynx. J Clin Oncol 1998;16:2514 21.
13. Leong SS, Wee J, Tay MH. Paclitaxel, carboplatin, and gemcitabine in
metastatic nasopharyngeal carcinoma: a Phase II trial using a triplet
combination. Cancer 2005;103:569 75.
14. Dugan M, Choy D, Ngai A, et al. Multicenter phase II trial of
mitoxantrone in patients with advanced nasopharyngeal carcinoma in
Southeast Asia: an Asian-Oceanian Clinical Oncology Association
Group study. J Clin Oncol 1993;11:70 6.
15. Foo KF, Tan EH, Leong SS, et al. Gemcitabine in metastatic
nasopharyngeal carcinoma of the undifferentiated type. Ann Oncol
2002;13:1506.
16. Poon D, Chowbay P, Cheung YB, et al. Phase II study of irinotecan
(CPT-11) as salvage therapy for advanced nasopharyngeal carcinoma.
Cancer 2005;103:57681.

Downloaded from http://jjco.oxfordjournals.org/ at Kainan University on March 24, 2015

Authors