To define the incidence and impact of tearing in patients receiving adjuvant docetaxel-based chemotherapy and assess for lacrimal duct obstruction (LDO) as a causative factor.

Consecutive patients with early breast cancer recommended for docetaxel-based chemotherapy with no prior ocular symptoms were included. Before and after completion of chemotherapy, patients underwent lacrimal drainage evaluation by computed tomographic dacrocystography (CT-DCG) and ophthalmic assessment. Eye symptoms were assessed at baseline, during, and after completion of chemotherapy.

Over a 22-month period, 100 patients were recruited. Asymptomatic LDO was present at baseline in 17% and 18% of patients, as assessed by ophthalmic review and CT-DCG, respectively. Overall, 86% of patients developed tearing, with no significant difference between those who did and did not have LDO (94% v 84%; P = .45). Blepharitis occurred in 37% and minor corneal epitheliopathy in 22% of patients, with neither condition predicting for the development of tearing. Impairment of visual activities was greatest after cycle one (70% of patients) but had decreased to < 5% by 4 months after treatment.

Tearing occurs in the majority of patients receiving adjuvant docetaxel-based chemotherapy regimens and occurred similarly in patients with and without LDO. There was poor concordance between CT-DCG and ophthalmic examination in the detection of LDO. Tearing and other eye symptoms impaired visual activities, but in nearly all patients, both symptoms and functional impairment were mild and had resolved by 4 months after chemotherapy. Our study demonstrates docetaxel-related tearing is not caused by LDO, and as such, evaluation or stenting of the duct is not considered necessary.

Women receiving adjuvant chemotherapy for breast cancer frequently report ocular symptoms during therapy. These include eye irritation and discomfort, blurred vision, accommodation disturbance, photophobia, conjunctivitis, hyperemia, and excessive tearing.1,2 Tearing has been reported to occur in up to 27% of women receiving systemic fluorouracil and 64% of women receiving docetaxel-based chemotherapy.3,4,5 It has been suggested that tearing may result from canalicular and nasolacrimal duct stenosis, and in severe cases, permanent sclerosing canaliculitis has been a reported cause.6,7,8 Given that lacrimal obstruction has been considered the primary cause of tearing, lacrimal duct stenting has been suggested as a treatment in selected patients.5,9

The use of taxane-based (paclitaxel and docetaxel) regimens in the adjuvant setting has been shown to improve survival in both node-positive and -negative patients.10,11 This has led to the increasing use of docetaxel-based chemotherapy regimens, which will likely lead to more women experiencing ocular symptoms such as tearing, with its resulting impact on quality of life. The incidence of tearing, the severity and duration of symptoms, and the proportion of cases related to true lacrimal obstruction in patients receiving adjuvant chemotherapy for early-stage breast cancer are not known. None of the major phase III studies examining the role of docetaxel in early breast cancer have described the rate of ophthalmologic symptoms experienced by patients receiving docetaxel. The purpose of this study was to define the incidence and impact of tearing in patients undergoing adjuvant docetaxel therapy and to evaluate the role of lacrimal duct pathology as an underlying causative factor.

Consecutive patients with early-stage breast cancer who were recommended for standard adjuvant docetaxel-based chemotherapy were recruited between January 2008 and October 2009 from two cancer centers. Eligible patients were required to not have had ocular symptoms (tearing, stinging, burning, itchiness, blurring, or altered sensation in the eye) in the 28 days before chemotherapy, to have a normal ophthalmic or radiologic evaluation of the nasolacrimal apparatus, to go on to receive a minimum of one cycle of chemotherapy, and to attend at least one follow-up eye review. Patients found to have asymptomatic nasolacrimal obstruction as assessed by both computed tomographic dacrocystography (CT-DCG) and lacrimal duct syringing were excluded.

All patients received dexamethasone premedication in line with standard clinical practice (ie, 4 mg orally for five to six doses commencing the day before treatment). The dose of each chemotherapy agent was based on actual body-surface area and in keeping with the standard dosage recommended for a given regimen.

The study was approved by the research and ethics committees of the Mount and Epworth-Freemasons Hospitals in Perth and Melbourne, respectively. All study procedures were performed in accordance with the Declaration of Helsinki.

Patient Evaluation

In the 28 days before commencing chemotherapy, patients underwent ophthalmic assessment by the study ophthalmologists. Assessment consisted of visual acuity and intraocular pressure measurements; slit-lamp examination for punctal stenosis, punctal malposition, expressible mucocoele, blepharitis, and intraocular inflammation; fluorescein dye disappearance test and fluorescein staining of cornea; irrigation of the lacrimal duct with an estimate of the percentage of fluid reflux; and gentle probing of both canaliculi. CT-DCG was performed, and patients completed an eye symptom questionnaire (Table 1).


Table 1. Eye Questionnaire

Table 1. Eye Questionnaire

Symptom Degree
Tearing No tearing
Mild tearing requiring occasional wiping of tears
Moderate tearing requiring wiping of tears every day
Severe tearing requiring wiping tears most of awake hours or limiting ability to read, drive, or carry out activities of daily living
Stinging, burning, altered eye sensation, itchy, or blurring A. Absent or present
B.    If present, symptom causes you to:
    a)    Stop driving: yes/no
    b)    Stop reading: yes/no
    c)    Stop other visual task: yes/no
    d)    Stop any activity: yes/no
    e)    Wipe one/both eyes: yes/no
C.    If present, it occurs:
    a)    > once per day
    b)    < once per day and > once per week
    c)    < once per week

Using the same questionnaire, eye symptoms were assessed on day 1 of each cycle of chemotherapy, at 21 to 28 days after the last cycle (defined as end of chemotherapy [EOC]), and at the follow-up visit (which occurred approximately 4 months after EOC). The severity of specific eye symptoms was graded using the Common Terminology Criteria for Adverse Events (version 3.0). Ophthalmic assessments were performed by the same ophthalmologist after completion of 50% of the planned chemotherapy and again within 21 days of the last cycle; a second CT-DCG was performed at 4 months after the last cycle of chemotherapy.


Two drops of the contrast agent (Omnipaque 300; GE Healthcare, Cork, Ireland) were administered to each eye 2 to 3 minutes before CT scanning. Patients were instructed to gaze upward during the scan to minimize radiation exposure to the cornea (corneal shields were not used so as to avoid artifactual distortion of CT images). Obstruction was classified as: 1) nil present, 2) partial obstruction, or 3) complete obstruction, with the site of obstruction defined as proximal, middle, or distal third. The same radiologist was responsible for reporting CT-DCGs within each center throughout the study.

Statistical Considerations

Given that this was an observational, prospective study, no formal sample-size calculation was performed. Results are presented as a proportion of the eligible population, and an exploratory analysis of the impact of total dose of docetaxel on tearing incidence, duration, and severity was undertaken. A weighted kappa test was used to assess for agreement between ophthalmic and CT-DCG evaluation of lacrimal duct patency and site of obstruction.

Symptom Management

Patients who developed tearing were advised to use commercial-brand artificial tears. In the presence of blepharitis, symptomatic relief with a warm towel compress placed against the closed eye was used, with or without polyethylene glycol 0.3% lubricant. In the event of conjunctivitis, chloramphenicol 0.5% drops and ointment were prescribed.

Over 22 months, 102 patients were recruited to the study; 100 patients received the planned treatment, were evaluable, and constituted the study population. Two patients were excluded (both withdrew consent after baseline evaluation). The mean age of patients was 52 years (range, 33 to 76 years), with 51% being pre- or perimenopausal and 49% postmenopausal. The majority of patients had stage II or III disease (stage I, 15%; II, 72%; III, 13%). Characteristics of the treatment administered are listed in Table 2. Fourteen patients reported having a history of allergic rhinitis but did not experience tearing in the 28-day period leading up to commencement of chemotherapy.


Table 2. Treatment Characteristics

Table 2. Treatment Characteristics

Chemotherapy No. of Patients Tearing (%)* Cumulative Docetaxel Dose (mg)
Mean Range
TAC 55 86 770 485-960
TCH 17 88 770 625-900
FEC-T 14 100 517 450-600
TCy 64
    + H 2 480 440-520
    − H 12 530 440-580

Abbreviations: FEC-T, fluorouracil 500 mg/m2, epirubicin 100 mg/m2, and cyclophosphamide 500 mg/m2 for three cycles followed by docetaxel 100 mg/m2 for three cycles; TAC, docetaxel 75 mg/m2, doxorubicin 60 mg/m2, and cyclophosphamide 600 mg/m2 for six cycles; TCH, docetaxel 75 mg/m2, carboplatin area under curve 6, trastuzumab 8 mg/kg loading, 6 mg/kg maintenance, for six cycles followed by trastuzumab for 17 cycles; TCy ± H, docetaxel 75 mg/m2 and cyclophosphamide 600 mg/m2 for four cycles with or without trastuzumab 8 mg/kg loading, 6 mg/kg maintenance, for four cycles followed by trastuzumab for 17 cycles.

*Proportion of patients experiencing tearing at some time during chemotherapy up to end of chemotherapy.

Incidence of Tearing

Overall, 86% of patients developed tearing at some point during the course of adjuvant chemotherapy, with the majority of patients (81.2%) experiencing tearing by cycle three (data not shown). The incidence of tearing according to chemotherapy regimen is shown in Figure 1. Onset of tearing according to cycle of treatment was 39% at day 1 of cycle two, 30% at cycle three, 12% at cycle four, 5% at cycle five, and 9% at cycle six; 5% of patients noted tearing for the first time at EOC. Patients receiving the FEC-T (fluorouracil, epirubicin, cyclophosphamide, and docetaxel) regimen received only three cycles of docetaxel, but all 14 patients (100%) experienced tearing, compared with ten (71.4%) of 14 patients receiving a four-cycle regimen and 62 (86.1%) of 72 patients receiving six cycles of docetaxel (Fig 2). Six patients (6%) reported grade 3 tearing at some time during their chemotherapy, with complete resolution in four patients and improvement to grade 1 in one patient at the postchemotherapy follow-up visit. The remaining patient, who reported grade 3 tearing at the time of follow-up, had patent lacrimal ducts on both postchemotherapy CT-DCG and ophthalmic examination.

Assessment of Lacrimal Drainage Obstruction

At baseline, some degree of lacrimal duct obstruction (LDO) was seen in 17 patients on ophthalmic review and in 18 patients on CT-DCG. There was poor agreement between both methods of lacrimal duct evaluation in any one patient, with a kappa agreement of 0.004 (95% CI, −0.199 to 0.191; Fig A1, online only). Patient compliance with the postchemotherapy CT-DCG and ophthalmic review was poor, with approximately 25% of patients failing to attend. In those who underwent these assessments, 30% and 22% of patients were assessed as having LDO, respectively. Of the 73 patients who underwent both ophthalmic review and CT-DCG, there was an increase in the level of agreement between the two techniques, but this remained only fair (kappa, 0.23; 95% CI, 0.009 to 0.463; Table 3).


Table 3. CT-DCG and Ophthalmic Examination

Table 3. CT-DCG and Ophthalmic Examination

CT-DCG Ophthalmic Examination (No. of patients)
Obstruction No Obstruction
    Obstruction 3 15
    No obstruction 14 68
End of chemotherapy
    Obstruction 7 15
    No obstruction 6 45

Abbreviation: CT-DCG, computed tomographic dacrocystography.

Patients were assessed for an association between LDO and the development of tearing. The presence of obstruction on CT-DCG at baseline was associated with the development of tearing in 94% of patients, but this was not significantly different from the 84% of patients with patent lacrimal ducts on CT-DCG who also went on to develop tearing (P = .45).

With respect to baseline ophthalmic examination, punctal stenosis was present in 27% of patients, and although 89% of these patients developed tearing, this was not significantly different from the 85% of patients who developed tearing without evidence of punctal stenosis at initial review. Irrigation of the lacrimal duct demonstrated obstruction in 17% of patients, with 94% of these patients subsequently experiencing tearing. Again, despite patency of the lacrimal duct on irrigation at baseline in the remaining patients, 84% experienced tearing (Appendix Fig A2, online only).

In six patients (6%), nasolacrimal irrigation worsened over the course of the study, with one patient developing complete obstruction. Nine patients (8.8%) had worsening of obstruction on CT-DCG, with complete obstruction occurring in eight. One third of patients who developed worsening irrigation continued to experience tearing at EOC, which was identical to the proportion of patients experiencing tearing despite no change of irrigation findings at that time. In contrast, 89% of patients who had worsening lacrimal obstruction demonstrated on CT-DCG continued to experience tearing, compared with 67% of those with unaltered CT-DCG findings.

Ophthalmic Findings

Blepharitis was present in 37% of patients at baseline, with an increase to 42% at EOC. Minor corneal epitheliopathy by fluorescein was identified in 18% and 22% of patients at baseline and end of chemotherapy, respectively. The presence of these ocular abnormalities did not correlate with the development or severity of tearing (data not shown).

Impact of Eye Symptoms on Daily Functioning

A range of other eye symptoms were experienced and reported by patients during the course of treatment, but they were generally mild and transient in nature. The following eye symptoms were reported by patients at some time point during chemotherapy delivery: itchiness, 59%; altered sensation, 58%; stinging, 40%; blurring, 63%; and burning, 18%. Over time, patients experiencing tearing more commonly complained of difficulties with reading than other aspects of daily activities such as driving or visual tasks. The highest rates of impaired daily visual functioning were reported after cycle one (70% of patients reported some degree of impairment), irrespective of the regimen administered. By EOC, the overall incidence was 25%, with only 4% of patients reporting impaired visual tasks at 4 months after treatment. Patients were less likely to complain of interference with several activities of daily living despite a gradual increase in the incidence of tearing over time (Fig 3).

Tearing has been reported after the use of docetaxel, with several case reports and retrospective studies demonstrating punctal stenosis or LDO as an underlying cause.12,13 A retrospective study of 148 patients (98% with metastatic disease) receiving docetaxel concluded that weekly administration of docetaxel was associated with the highest rates of LDO and canaliculi obstruction. The largest prospective study of tearing in patients with breast cancer was conducted in 56 women with metastatic disease. After mean intervals of 2 and 3 months, 64% and 39% of patients developed bilateral tearing with docetaxel weekly or every 3 weeks, respectively. The incidence of canalicular stenosis before initiation of docetaxel administration was not specified, but 50% and 18% of those receiving treatment weekly or every 3 weeks, respectively, had worsening canalicular stenosis despite conservative eye management. Surgical stenting was undertaken in eight patients, with improvement or resolution of tearing in all.9

In the adjuvant setting, the incidence of tearing and other eye symptoms associated with docetaxel-based treatment is infrequently reported. The lack of published data regarding tearing incidence may result in part from the low rates of grade 3 or 4 events, although in the E1199 study, incidences of < 1% and 5% were reported in patients receiving docetaxel every 3 weeks or weekly, respectively.14,15,16 None of the other major phase III studies conducted to examine the role of docetaxel in the adjuvant setting listed ophthalmologic symptoms in the list of docetaxel-related toxicities, and tearing was not mentioned as a specific entity. However, in the docetaxel (Taxotere; sanofi-aventis, Bridgewater, NJ) package insert, mention of ophthalmologic toxicities including tearing is made under the heading of postmarketing experiences, and reference is also made in the patient information leaflet.

To our knowledge, this is the first prospective study to evaluate the incidence of tearing and other eye symptoms in the early breast cancer setting. Furthermore, we used two methods to assess for an association between LDO and tearing. We have shown that evidence of LDO before commencement of chemotherapy is present in up to 18% of asymptomatic patients. More importantly, patients developed tearing regardless of the presence of obstruction before or after completion of chemotherapy. Up to 42% of patients had blepharitis, and 22% of patients had minor corneal epitheliopathy identified after chemotherapy. Again, these findings were not significantly associated with the development of tearing. Our study was not designed to evaluate specific causes of tearing. However, it is possible that other known causes of hypersecretion such as lacrimal gland irritation by cytotoxic drugs, subclinical corneal stimuli as a result of changes in pH or osmolarity, or chemical triggering of the corneal nerves may be involved.

Despite 86% of patients experiencing tearing during adjuvant chemotherapy, 80% reported this symptom as being mild or moderate. By 4 months after the last cycle of chemotherapy, 70% of patients reported no tearing, 29% reported mild intermittent tearing, and only one patient complained of ongoing severe symptoms. There was a trend for lower rates of tearing in patients receiving lower cumulative doses of docetaxel, except for patients who received the FEC-T regimen with fluorouracil before docetaxel, which is in keeping with the known effects of fluorouracil.

It is beyond the scope of this study to explain the poor concordance between CT-DCG and irrigation results in the assessment of lacrimal obstruction. However, it should be noted that there exist fundamental differences between the two techniques. CT-DCG involves passive transition of a constant volume of dye placed in the eye, as compared with gentle positive pressure placed during saline syringing. Thus, it is possible that incomplete LDO may have been missed by CT-DCG. Furthermore, it was not possible to control for subtle differences in the dwell time of dye instillation and subsequent radiologic review that may have occurred during CT-DCG, whereas the evaluation of the lacrimal duct at the time of ophthalmologic examination was more likely to be consistent. The lack of correlation between the presence of LDO and the development of tearing in the current study suggests that formal duct evaluation pretreatment is not necessary. It is possible that regular ophthalmic assessment, as was undertaken in this study, may have led to earlier intervention with conservative treatments, although this was not formally assessed. We have demonstrated that LDO is not the primary cause of tearing and that symptoms are generally short lived. As such, symptomatic improvement may be achieved with conservative approaches such as the placement of a hot compress over the closed eyelid, gentle vertical massaging of the eyelids, and regular use of isoosmolar lubricant drops.

In conclusion, we have shown that docetaxel, a component of several standard adjuvant chemotherapy regimens, causes a high rate of tearing and that the high incidence of this adverse effect has an impact on several aspects of daily living. However, patients receiving docetaxel-based treatment with or without fluorouracil should be reassured that tearing is generally mild and transient in the majority of cases, and surgical intervention is not required. The findings from our study do not support the need for stenting of the lacrimal duct before commencement of chemotherapy. Ophthalmologic assessment during treatment may be recommended for some patients, if tearing is severe, to exclude blepharitis and corneal abnormalities.

© 2013 by American Society of Clinical Oncology

See accompanying editorial on page 2076

Supported by Study Grant No. BCRC108 from sanofi-aventis Australia.

Presented at the 32nd Annual San Antonio Breast Cancer Symposium, San Antonio, TX, December 9-13, 2009.

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: None Consultant or Advisory Role: Arlene Chan, sanofi-aventis (C); Richard H. De Boer, sanofi-aventis (C) Stock Ownership: None Honoraria: Arlene Chan, sanofi-aventis; Richard H. De Boer, sanofi-aventis Research Funding: Arlene Chan, sanofi-aventis Expert Testimony: None Other Remuneration: None

Conception and design: Arlene Chan, Charles Su

Provision of study materials or patients: Arlene Chan, Richard H. De Boer

Collection and assembly of data: All authors

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

1. JD Griffin, MB Garnick: Eye toxicity of cancer chemotherapy: A review of the literature Cancer 48: 15391549,1981 Crossref, MedlineGoogle Scholar
2. M Vizel, MW Oster: Ocular side effects of cancer chemotherapy Cancer 49: 19992002,1982 Crossref, MedlineGoogle Scholar
3. AS Eiseman, JC Flanagan, AB Brooks , etal: Ocular surface, ocular adnexal and lacrimal complications associated with the use of systemic 5-fluorouracil Ophthal Plast Reconstr Surg 19: 216224,2003 Crossref, MedlineGoogle Scholar
4. A Stevens, D Spooner: Lacrimal duct stenosis and other ocular toxicity associated with adjuvant cyclophosphamide, methotrexate and 5-fluorouracil combination chemotherapy for early stage breast cancer Clin Oncol (R Coll Radiol) 13: 438440,2001 MedlineGoogle Scholar
5. B Esmaeli, L Hidaji, RB Adinin , etal: Blockage of the lacrimal drainage apparatus as a side effect of docetaxel therapy Cancer 98: 504507,2003 Crossref, MedlineGoogle Scholar
6. V Lee, CR Bentley, JM Olver: Sclerosing canaliculitis after 5-fluorouracil breast cancer chemotherapy Eye (Lond) 12: 343349,1998 Crossref, MedlineGoogle Scholar
7. A Hassan, JJ Hurwitz, RL Burkes: Epiphora in patients receiving systemic 5-fluorouracil therapy Can J Ophthalmol 33: 1419,1998 MedlineGoogle Scholar
8. HM Brink, LV Beex: Punctal and canalicular stenosis associated with systemic fluorouracil therapy: Report of five cases and review of the literature Doc Ophthalmol 90: 16,1995 Crossref, MedlineGoogle Scholar
9. B Esmaeli, S Amin, V Valero , etal: Prospective study of incidence and severity of epiphora and canalicular stenosis in patients with metastatic breast cancer receiving docetaxel J Clin Oncol 24: 36193622,2006 LinkGoogle Scholar
10. Laurentiis M De, G Cancello, D D'agostino , etal: Taxane-based combinations as adjuvant chemotherapy of early breast cancer: A meta-analysis of randomized trials J Clin Oncol 26: 4453,2008 LinkGoogle Scholar
11. M Martín, M Seguí, A Antón , etal: Adjuvant docetaxel for high-risk, node-negative breast cancer N Engl J Med 363: 22002210,2010 Crossref, MedlineGoogle Scholar
12. DW Spell, FF Estephan, JT Lin , etal: Epiphora induced by intermittent docetaxel (Taxotere) in patients with non-small cell lung cancer Cancer Invest 21: 550552,2003 Crossref, MedlineGoogle Scholar
13. PE Kintzel, LB Michaud, MK Lange , etal: Docetaxel-associated epiphora Pharmacotherapy 26: 853867,2006 Crossref, MedlineGoogle Scholar
14. M Martin, T Pienkowski, J Mackey , etal: Adjuvant docetaxel for node-positive breast cancer N Engl J Med 352: 23022313,2005 Crossref, MedlineGoogle Scholar
15. W Eiermann, T Pienkowski, J Crown , etal: Phase III study of doxorubicin/cyclophosphamide with concomitant versus sequential docetaxel as adjuvant treatment in patients with human epidermal growth factor receptor 2–normal, node-positive breast cancer: BCIRG-005 trial J Clin Oncol 29: 38773884,2011 LinkGoogle Scholar
16. JA Sparano, M Wang, S Martino , etal: Weekly paclitaxel in the adjuvant treatment of breast cancer N Engl J Med 358: 16631671,2008 Crossref, MedlineGoogle Scholar


We thank Andy Whyte, MD, and Mark Scott, MD, for radiologic assessment.


No companion articles


DOI: 10.1200/JCO.2012.45.6574 Journal of Clinical Oncology 31, no. 17 (June 10, 2013) 2123-2127.

Published online May 06, 2013.

PMID: 23650421

ASCO Career Center