Oral Pilocarpine for Post-Irradiation Xerostomia in Patients with Head and Neck Cancer

Patients

Eligible patients had to have received at least 4000 cGy of external-beam radiation therapy for head and neck cancer more than four months before entry into the study. All the patients had clinically important xerostomia. They all had at least one parotid gland and some evidence of residual salivary function on physical examination. Primary reasons for exclusion from the study included clinically important uncontrolled cardiac, renal, and pulmonary disease, ocular disease, and other chronic diseases that could potentially interfere with the evaluation of the safety and efficacy of pilocarpine. For example, patients were excluded if they required tricyclic antidepressants or antihistamines with anticholinergic effects, beta-blockers, or pilocarpine for ophthalmic indications. Before entry, the patients underwent screening for medical eligibility, including a history taking, physical examination, 12-lead electrocardiography, laboratory screening, and ophthalmologic examination. The laboratory examination included a complete blood count, serum chemical profile, and serologic tests for hepatitis B surface antigen. Informed consent and approval from the institutional review board were obtained. The study was conducted in double-blind fashion, and the code was not broken until the study was completed.

Treatment Protocol

Patients were assigned randomly (with computer-generated randomization codes with a block size of six) to receive tablets containing 5 mg of pilocarpine, 10 mg of pilocarpine, or placebo; the tablets were administered with water three times a day at mealtimes. The duration of treatment was 12 weeks. Patients were seen before the start of treatment and at four-week intervals. Compliance with the dosing schedule was ascertained by counting tablets at each monthly office visit. At each scheduled visit a subjective assessment of efficacy was undertaken with visual-analogue scales. The patients answered questions about their sensations of oral dryness and oral comfort and their ability to speak, chew and swallow, sleep, and wear dentures. They also answered a general question about the overall condition of their xerostomia.

Salivary function was assessed by measuring salivary flow twice at each visit. The first collection of saliva occurred before the dose of medication, and the second 60 minutes after medication had been given. The patients were allowed no gustatory stimulation for at least 60 minutes before the first saliva collection. In each collection, specimens were obtained in the following order: unstimulated whole saliva, unstimulated parotid saliva, and stimulated parotid saliva. Unstimulated whole saliva was defined as the amount of saliva produced for one minute without the patient's swallowing. The production of saliva by the parotid gland was measured by placing a Carlson-Crittenden cup over the parotid papillae and collecting the saliva for one minute⁹. After stimulation of the dorsolateral tongue surface with a 2 percent citrate solution applied at 30-second intervals for 2 minutes, the final sialometric specimen was collected. If no parotid flow was measured within 10 minutes, by either the stimulated or the unstimulated method, production by that measure was considered to be zero. All the specimens were weighed to an accuracy of within 0.0001 g.

An adverse effect was defined as any unintended effect or abnormal clinical observation. The patients recorded their impressions daily in diaries that were reviewed at each visit, and they were questioned in order to elicit unbiased responses about adverse effects. The patients who withdrew from the study because of adverse effects were followed by the investigator until the adverse effect had resolved. In addition to routine assessment, all the patients underwent an ophthalmologic examination, updated history taking, physical examination, and 12-lead electrocardiography at the conclusion of the study or at the time of withdrawal.

Analyses of Salivary Function and Questionnaires for Patients

Approximately 35 percent of the patients had no measurable whole saliva at base line. A patient who had an increase in saliva production at any visit after beginning the study treatment was considered to have had a response. A 100-mm visual-analogue scale was used to record the responses to each of six questions. The scale was set up with negative responses on the left and positive responses on the right. The patients marked their responses on the scale in relation to these extremes. In each analysis, the score at base line (visit 1) was subtracted from the later scores to assess change. A patient with an increase of 25 mm or more from base line in the score on the visual-analogue scale was classified as having responded. Nine patients (two assigned to placebo, six to 5 mg of pilocarpine, and one to 10 mg of pilocarpine) who had base-line scores of 75 mm or higher for oral dryness on the visual-analogue scale were excluded from the analysis that required a base-line measurement before treatment. The patients were also asked three categorical questions about whether or not their oral comfort, oral dryness, and ability to speak had improved “immediately” after each dose of medication.

Statistical Analysis

The primary measures of efficacy were based on the proportion of patients with a response who had an increase from base line of at least 25 mm on the assessment of oral dryness, as defined before the unblinding of the study, and by any increase from base line in the amount of saliva collected. Overall improvement and the use of agents to provide oral comfort were evaluated on the basis of three categories of responses (improved, no change, or worsened). For the purpose of analysis, the flow from both parotid glands was added together, and the data were presented as the total parotid flow. A chi-square test was used to compare the differences between treatment groups in the proportion of responses. The results of three-way chi-square tests comparing the three study groups are reported, except where one of the pilocarpine groups is compared with placebo, as stated in the text.

Patients for whom any data were obtained after treatment were included in the intention-to-treat analysis. The main emphasis of the study was on this analysis. A second group of patients who could be evaluated was also defined, composed of patients who took their tablets with a compliance of 75 percent or higher, who did not miss more than four consecutive days of dosing, and who otherwise complied with the protocol. This second group also included patients who discontinued the study drug because of adverse effects. Ability to be evaluated was determined for each visit separately, so that patients for whom data could not be evaluated at one visit may have had data that could be evaluated at another visit. Data for the “final visit” represent the last assessment made for each patient. This was visit 4 for patients who completed 12 weeks of treatment and an earlier visit for those who withdrew from the study.

All the treated patients were included in the analysis of safety.

Table 1. Table 1. Base-Line Characteristics of the Patients with Post-irradiation Xerostomia, According to Treatment Group.

Of a total of 207 patients enrolled at 39 treatment sites, 65 patients were randomly assigned to placebo, 73 to treatment with 5 mg of pilocarpine, and 69 to treatment with 10 mg of pilocarpine. The base-line characteristics of the three groups are shown in Table 1. All the patients had received radiation to some portion of the salivary apparatus in doses totaling at least 4000 cGy. The doses of radiation ranged from 4001 to 7500 cGy. No significant base-line differences in salivary function were identified between groups (data not shown).

Table 2. Table 2. Incidence of Withdrawal from the Study and Study Completion, According to Treatment Group.

One hundred sixty-six patients completed 12 weeks of study treatment as scheduled. Forty-one patients withdrew from treatment before completing the study. The reasons for early withdrawal are shown in Table 2. Data were obtained after the first dose of the study drug for 193 patients. Because of missing base-line data for 2 patients, the analyses requiring such data for comparison included 191 patients. This group was defined as the intention-to-treat group.

Clinical Outcomes

Table 3. Table 3. Effect of Oral Pilocarpine on Symptoms of Post-irradiation Xerostomia.

Dryness of the mouth is the primary symptom of xerostomia. In both pilocarpine groups there was statistically significant improvement in feelings of oral dryness and mouth comfort and an overall improvement of xerostomia after treatment (Table 3). There was nonsignificant improvement in sleeping, chewing and swallowing food, and denture comfort.

A significantly higher proportion of patients in the two pilocarpine groups had improvement in the sensation of intraoral dryness in the intention-to-treat analysis (P = 0.037). Oral pilocarpine resulted in relief of symptoms in every area assessed (Table 3). Both the 5-mg group and the 10-mg group had higher proportions of responses than the placebo group (P = 0.027 and P = 0.020, respectively). Among the patients who could be evaluated, there was a significant difference between the two pilocarpine groups and the placebo group (P = 0.010) in the proportion of patients who had improvement in the sensation of oral dryness. Pairwise comparisons showed that the 5-mg and 10-mg groups differed significantly from the placebo group (P = 0.004 and P = 0.018, respectively).

A general question was used to assess each patient's condition on the basis of recollections of the severity of symptoms before treatment began. This assessment measured each patient's sense of improvement on the basis of the patient's recall of severity at base line. In the intention-to-treat cohort there was a significant improvement in the treatment groups with respect to overall assessment scores (P = 0.010). Pairwise comparisons showed that the 5-mg group improved significantly more than the placebo group (P = 0.003), whereas improvement in the 10-mg group was not significant (P = 0.052). The analysis of the patients who could be evaluated (Table 3) showed that the proportion who had a response overall differed significantly between treatment groups (P = 0.016). Pairwise comparisons showed that the 5-mg group differed significantly from the placebo group (P = 0.003), whereas the 10-mg group did not (P = 0.074).

Figure 1. Figure 1. Improvement in the Sensation of Oral Dryness and Overall Improvement in Xerostomia, as Reported by the Patients in the Intention-to-Treat Cohort at Each Visit during the 12 Weeks of Treatment.

There was an apparent time-dependent and drug-related benefit in the case of the subjective end points. The percentage of patients who responded to pilocarpine therapy with symptomatic improvement in oral dryness (Figure 1), comfort, speaking, and other effects continued to increase throughout the study, with the best responses generally occurring at visit 4.

The ability to speak without requiring liquids was improved, but not significantly (P = 0.056), in the pilocarpine groups in the intention-to-treat analysis (Table 3). Pairwise comparisons showed that the 5-mg and 10-mg groups were both statistically different from the placebo group (P = 0.037 and P = 0.028, respectively). The analysis of patients who could be evaluated revealed similar trends (P = 0.015). The proportion of patients who had responses at a dose of 10 mg was no higher than the proportion who had responses at a dose of 5 mg.

There was significantly improved comfort of the mouth and tongue in the pilocarpine-treated subjects in the intention-to-treat group (P = 0.001) (Table 3). The 5-mg group and the 10-mg group were both significantly different from the placebo group (P ≤ 0.002 for both). The analysis of patients who could be evaluated showed similar results (P = 0.001), with the 5-mg and 10-mg groups both differing statistically from the placebo group (P ≤ 0.001 for both).

The patients were asked whether they used various methods of improving their oral comfort (such as eating hard candy, using saliva substitutes or sprays, chewing gum, sipping water, and brushing their teeth) more often than at base line, about the same, or less often (Table 3). The intention-to-treat analysis did not find significantly more improvement in this area with pilocarpine (P = 0.075), but the analysis of patients who could be evaluated found a statistically significant difference between the two pilocarpine groups and the placebo group (P = 0.009) in the use of agents for oral comfort.

There was significant improvement among the patients who could be evaluated (generally P<0.01) in both the 5-mg and the 10-mg treatment groups with respect to dryness, oral comfort, and ability to speak immediately after receiving a dose of the study drug (data not shown).

Production of Whole Saliva and Parotid Saliva

Table 4. Table 4. Effect of Oral Pilocarpine on the Production of Whole Saliva and Unstimulated Parotid Saliva as Measured Every 4 Weeks over a 12-Week Study Period in the Intention-to-Treat Analysis.

When the collections of whole saliva made after each dose were compared with those obtained before the dose in the intention-to-treat group, there was a difference between treatment groups (Table 4). After the first dose during visit 1, there were increases of borderline significance in the production of whole saliva in the intention-to-treat analysis (P = 0.058). There was significant improvement in this value at visit 2 (P = 0.004) and visit 3 (P = 0.043), but not at visit 4. The results at the final visit showed an increased proportion of responses in the 10-mg group as compared with the placebo group, although the difference was not significant (P = 0.129). The analysis of patients who could be evaluated showed similar results.

In the intention-to-treat analysis of unstimulated production of parotid saliva, a significantly higher proportion of patients in the 5-mg and 10-mg groups than in the placebo group responded to a dose at visits 1, 2, and 3 (Table 4). There was no statistically significant difference between treatments at visit 4 or at the final visit. The group of patients who could be evaluated yielded similar results.

Adverse Effects

Table 5. Table 5. Incidence of Adverse Effects.

Adverse effects were generally mild and occasionally of moderate severity. Three patients in the 10-mg group had sweating that was considered severe. Twenty-six patients withdrew from the study because of adverse effects, mostly sweating: 2 patients in the placebo group (3.1 percent), 4 in the 5-mg group (5.5 percent), and 20 in the 10-mg group (29 percent). The most frequent adverse effects are summarized in Table 5. There were no serious drug-related adverse effects in this study.

The results of this study show that pilocarpine can improve many symptoms of radiation-induced xerostomia. The percentage of patients who had improvement in symptoms continued to increase throughout the study, with the best responses generally occurring at 12 weeks. The reasons for a time-related improvement are unknown but are probably related to changes in the oral mucosa as pilocarpine treatment increased the amount of saliva produced. This time-dependent phenomenon was observed in a previous study⁷.

Although subjective improvement was consistently observed with pilocarpine therapy, the objective sialometric findings were less clear. Increased salivary flow was noted in the patients treated with pilocarpine, but the effect was not sustained and it varied throughout the treatment period. The disparity between subjective and objective findings may be due to physiologic or methodologic constraints. The sialometric methods used in this study were adapted from techniques of collection developed for persons producing higher volumes of saliva and may not have detected smaller increases in salivary flow. Although the actual volume of saliva may not correlate with the clinical response, small increases in the amount of saliva may be sufficient for clinical improvement. Patients with little or no production of saliva may derive major clinical benefits from minor increases in salivation, whereas patients with a greater initial salivary flow may not perceive the symptomatic relief produced by an equivalent increase in volume.

It is likely that there are both immediate and delayed effects of such an increase, and both may be clinically important. Salivary mucins contributed by the submandibular, sublingual, and minor salivary glands throughout the oral cavity are reported to provide protection in the mouth against chemical, infectious, and mechanical insults; to provide lubrication; and to prevent the desiccation of soft tissues¹⁰. Minor glands produce less than 10 percent of the total volume of saliva but contribute more than 70 percent of the total mucin content¹¹. Artificial preparations of saliva that contain mucin provide substantially more symptomatic relief to patients with xerostomia than do conventional non-mucin substitutes^12-16.

The literature clearly demonstrates that constituents of saliva such as mucins, proteins, glycoproteins, and electrolytes are stimulated by pilocarpine and other autonomic agonists¹⁷. Mandel and coworkers^18,19 documented increases in salivary amylase, protein-bound carbohydrate, lysozyme, total protein, and calcium after pilocarpine stimulation in normal subjects. Other reports document both time- and dose-related increases in mucin secretion after pilocarpine stimulation, with mucin secretions increasing in successive saliva collections^20-23. Any increase in salivation, no matter how small, that is accompanied by increases in the production of beneficial constituents may benefit patients with xerostomia. Quantitation of these constituents adhering to oral mucosa is probably not possible in patients with very small increases in salivary volume and may only be detected by measuring surrogate clinical responses. It is also possible that improvement in the oral mucosal membranes, measured on the basis of a decrease in symptoms, may require a long time to reverse atrophy or other changes brought about by the absence of saliva or its constituents.

In conclusion, pilocarpine can improve many symptoms associated with post-irradiation xerostomia, including oral dryness, discomfort, and difficulty in speaking; reduce the need for comfort agents; and lead to overall improvement. In this study, the treatment group that received 5 mg of pilocarpine three times daily had the best overall outcome with respect to saliva production and relief of symptoms of xerostomia, when both side effects and efficacy are considered.

In addition to the study authors, the following persons participated in the pilocarpine hydrochloride study: S.R. Andresen, the Carle Foundation, Carle Clinic; A. Brady, Porter Memorial Hospital; W. Carl, Roswell Park Memorial Institute; L. Davis, Philadelphia Veterans Affairs Medical Center; L.W. Davis, Montefiore Medical Center; R.L.S. Doggett, Radiological Associates of Sacramento; J. Dragovic, Henry Ford Hospital; T.M. Durham, University of Nebraska Medical Center; P. Eisenberg, Marin Oncology Associates; J.R. Eltringham, Swedish Hospital Tumor Institute; J.C. Frich, Jr., West Virginia University Health Sciences Center; Z. Khan, University of Louisville; T.A. Kiersch, Tucson, Ariz.; W.T. Knapp, St. Mary's Medical Center; P.B. Lockhart, Carolinas Medical Center; R.A. Lustig, Cooper Hospital University Medical Center; C.J. Meakin, the Christ Hospital Cancer Center; J. Messura, University of Iowa Hospitals; D. Peterson, University of Connecticut Health Center; N.L. Rhodus, University of Minnesota School of Dentistry; T.K. Robbins, University of California, San Diego, Medical Center; M. Roberts, Regional Oncology and Hematology Association; B. Rodu, University of Alabama at Birmingham; E. Root, Kelsey-Seybold Clinic; J. Schwade, University of Miami School of Medicine; J.Y. Suen, University of Arkansas for Medical Sciences; and S.G. Taylor IV, Rush-Presbyterian-St. Luke's Medical Center.