Evaluation of a pharmacist-led virtual thiopurine clinic

Introduction

Clinical thiopurine use

Autoimmune hepatitis (AIH) is a rare progressive immune-mediated disease of the liver affecting around 10–17 people per 100,000 in Europe^[1]
. Treatment involves long-term immunosuppression, with first-line treatment that is usually combination therapy with steroids and thiopurines (e.g. azathioprine and mercaptopurine [see below for information on their metabolism and clinical effects]), followed by thiopurine monotherapy, providing there is histopathological and serological evidence of disease remission^[1]
.

Inflammatory bowel disease (IBD), which also comprises Crohn’s disease (CD) and ulcerative colitis (UC), are chronic inflammatory gastrointestinal disorders affecting around 400 people per 100,000 in the UK. IBD significantly impairs quality of life and requires life-long management^[2]
. It is this management, particularly the therapeutic monitoring of thiopurines, that creates significant workloads for hospital gastroenterology departments. The average cost of each IBD patient to the NHS is £3,000 per year, with up to half of the total costs attributed to relapsing patients^[2]
.

Thiopurines represent second-line treatments for patients with UC and CD. The National Institute for Heath and Care Excellence (NICE) recommends initiating treatment when a patient experiences two or more inflammatory relapses in a 12-month period requiring steroid therapy, or those with steroid-dependent disease or treatment failure of aminosalicylites to maintain remission^[3],[4]
. Patients can be started on thiopurines in combination with biological treatment (e.g. infliximab) to minimise drug antibody development^[3],[4]
. Initiation of thiopurine therapy without regular clinical and biochemical evaluation exposes patients to sub-therapeutic dosing, and increases the risk of drug intolerance, poor compliance and myelosuppression^[5]
.

A higher immunosuppressant effect is required in patients with IBD to achieve disease remission and is reflected in the management pathway (the therapeutic target for IBD patients on azathioprine is 2–2.5mg/kg, and 1–1.25mg/kg for mercaptopurine monotherapy if thiopurine methyltransferase [TPMT] is normal). For patients with AIH, treatment is usually in combination with a low dose of prednisolone and azathioprine 1mg/kg. However, the target can be increased up to 2mg/kg when the disease is not controlled^[1],[5]
.

The incidence of adverse reactions associated with thiopurines is affected by the dose, duration of therapy, and concurrent medicine. The British Society of Gastroenterology guidelines for IBD recommend early intense monitoring and 12-weekly regular monitoring of blood indices, since thiopurines may induce myelotoxicity at any stage during therapy^[5]
. This is a dose-related and generally reversible drug effect that commonly features leucopoenia, thrombocytopenia and transaminitis^[6],[7]
. Myelosuppression can be aggravated in those patients with low or deficient levels of TPMT.

All patients must undergo screening of TPMT levels before starting therapy and the therapeutic target should be reviewed according to the range. If patients have a low level of TPMT, the target dose of thiopurines should only be 50% of the full dose; however, if the patient is deficient, thiopurines are not recommended^[5]
. Serial blood testing is required at weekly and then fortnightly intervals during the induction phase; however, this frequency may change according to blood results to minimise adverse events and prevent admissions^[5]
. Treatment failure with thiopurines relate to drug intolerance in around 30% of patients and hepatotoxicity in around 10% of patients^[8],[9]
.

Thiopurine metabolism and clinical effects

Azathioprine is a pro-drug that, after an oral dose, undergoes metabolism to mercaptopurine. Further metabolism occurs through competing pathways by three different enzymes: xanthine oxidase oxidizes mercaptopurine into the inactive thiouric acid; hypoxanthine phosphoribosyltrasferase converts mercaptopurine into thioinosine monophosphate, of which further intracellular metabolism results in the formation of the active metabolite 6-thioguanine nucleotide (TGN); thiopurine methyltransferase methylates mercaptopurine into the inactive methylmercaptopurine. TPMT also methylates thioinosine monophosphate into methylmercaptopurine nucleotides (see Figure). TGNs are the active product of thiopurine metabolism, with levels correlating with effective clinical response in IBD^[8]
.

Allopurinol, a xanthine oxidase inhibitor, is known to enhance the myelosuppression effect of thiopurines as it can ‘shunt’ drug metabolism from 6-methyl-mercaptopurine to the active metabolite TGN, to mitigate hepatotoxicity or intolerance-related side effects; therefore, lower thiopurine doses (between 25%–30% of the standard dose) are required^[8],[9],[10]
.

This interaction has been used in clinical practice as a benefit to mitigate the hepatotoxicity or intolerance effect induced by thiopurines when used as monotherapy.

Pharmacist involvement in the gastroenterology multidisciplinary team

Several audits, surveys and reports conducted between 2011 and 2013 highlighted differences in the level of involvement of pharmacists as part of the IBD multidisciplinary team (MDT).

An audit undertaken by the Royal College of Physicians (RCP) in 2011 of clinical IBD leads from 160 UK healthcare trusts and health boards found that a named pharmacist with an interest in gastroenterology was attached to the IBD team in less than 50% of the sites. The RCP audit report recommended that clinical pharmacy support for the IBD team should be strengthened, given the high cost and complexity of the drug regimens that are often used^[11]
.

The following year, a survey of acute trusts and health boards undertaken by the UK Clinical Pharmacy Association, the Royal Pharmaceutical Society and the University of Brighton found that 20% of NHS trusts had a part-time pharmacist working within the framework of an IBD MDT^[12]
. And, in 2013, the IBD Standard Group reported the lack of involvement of specialist IBD pharmacists in the management of biological therapies for IBD patients^[13],[14]
.

Following these recommendations, the role of the pharmacist has evolved over recent years to actively pursue sub-speciality interests and to embrace the challenge of therapeutic drug monitoring. The creation of a pharmacist-led thiopurine clinic with appropriate clinical governance was a logical and necessary response to the variation in care and is supported by published evidence from other clinical subspecialties^{[15],[16],[17]}
.

Barnsley Hospital NHS Foundation Trust

Barnsley Hospital NHS Foundation Trust is a general district hospital which covers around 250,000 people and supports about 1,400 IBD adult patients. The gastroenterology department participated in the 2020 national benchmarking survey and the IBD team includes gastroenterologist consultants, IBD nurses and an IBD pharmacist (6, 1.2 and 0.5 full time-equivalent, respectively)^[18]
.

The IBD pharmacist:

• Provides directed support to newly diagnosed patients and those admitted to the gastroenterology ward as part of the MDT, alongside gastroenterologists and specialist nurses;
• Cares for patients who are admitted on the gastroenterology ward;
• Screens elective admissions for regular biologic therapy;
• Provides advice and escalates therapy when required through the pharmacy telephone clinic or pharmacy helpline;
• Educates and counsels patients receiving ongoing care and treatment on how to achieve a good quality of life^[14]
  .

The IBD service model was created in October 2012 and evaluated in 2014, with the results of the evaluation study published as an abstract at the 23rd United European Gastroenterology Week (UEG) in 2015.

Over the past seven years the scope of the specialist IBD pharmacist role has expanded to integrate the monitoring of other immunosuppressants (e.g. mycophenolate, methotrexate and tofacitinib); the prescription of intravenous biologics (e.g. infliximab, vedolizumab and ustekinumab); and undertaking pre-assessment observations (e.g. biochemistry and clinical) prior to their next biologic infusion. This is in line with the current recommendations related to pharmacy support outlined in the UK IBD Standards^[19]
.

Aim

The aim of this study was to determine the clinical impact of a pharmacist-led, virtual, telephone thiopurine clinic on drug monitoring, safety and level of satisfaction related to the delivery of the service in a cohort of IBD and AIH patients. The study was conducted at Barnsley Hospital NHS Trust over a three-month period between 1 May and 31 July 2014. A follow-up survey was conducted five years later in December 2019 to re-assess patient satisfaction with the service.

Method

Study design

The study in 2014 was designed to examine:

1. Pharmacist-led, virtual, telephone medicine optimisation based on drug tolerability, biochemical surveillance, TGN analysis, and patient adherence;
2. Thiopurine adverse drug reactions (ADRs) monitoring;
3. Patient satisfaction of the pharmacist-led virtual clinic.

In 2019, patient satisfaction was re-assessed to determine whether there was any change over the past seven years.

Patient selection

To be recruited into the study, patients had to be aged 18 years or over, with a confirmed diagnosis of IBD or AIH. All patients were referred by the gastroenterologist to the pharmacist to initiate thiopurine therapy or to escalate their dose to the therapeutic target. Patients who had reached their therapeutic target and were stable at the time of study recruitment were excluded.

The 2019 re-evaluation of patient satisfaction included all patients with IBD taking thiopurines who were undergoing monitoring through the clinic.

Screening

All patients were screened to determine their TPMT enzyme level before starting treatment with thiopurines to guide their required therapeutic target range^[5]
. The empirical clinical therapeutic dose is based on the patient’s weight, as follows: azathioprine is 2–2.5mg/kg; mercaptopurine is 1–1.25mg/kg (with normal TPMT activity) and 50% therapeutic target dose for those patients who present low level of TPMT^[5]
. The therapeutic target in AIH for azathioprine is 1mg/kg with normal TPMT activity^[2]
.

Traditionally, gastroenterologists initiate thiopurines gradually, building tolerance up to the target dose, an approach that makes patients more likely to adhere to therapy. However, there is a delay before the therapeutic effects of thiopurines start and current guidelines recommend to start therapy on target as soon as possible to reduce the gap and achieve therapeutic effect at an early stage^[5]
. Nevertheless, thiopurines cannot induce remission and are normally introduced at the same time with a short course of steroids^[5]
. Both strategies could be acceptable if the target dose is reached quickly during the first month.

Routine blood tests (to determine TGN levels) for every patient is not cost-effective, so it is not recommended and should only be considered when poor compliance or drug intolerance is suspected, or to ensure therapeutic range in patients taking allopurinol. As drug intolerance and hepatotoxicity normally occur during induction, the dose can be tailored according to TGN results within the therapeutic range afterwards^[5]
.

Patient review 

Patients were referred by consultants for an initial face-to-face consultation with the pharmacist, to start immunosuppressant therapy. Patients were counselled on the monitoring plan required for thiopurines and also provided with an IBD thiopurines leaflet that contained this information and further details about the clinic. The nature of the study was carefully explained to each patient and consent secured before formal induction. Subsequent interactions with patients occurred in a virtual clinic environment that uses telephone, mobile text and email to follow up with patients. The telephone clinic includes three pharmacists (one specialist pharmacist [Band 7] and two Band 8a pharmacists) to operate the service the whole year round and cover leave.

Data collection

Data were collected prospectively for three consecutive months from 1 May to 31 July 2014 on:

• The number of routine blood tests and TGN assays requested;
• The frequency of thiopurine dose optimisations implemented by the pharmacist;
• The number of patients reaching therapeutic target;
• The frequency of attendance for phlebotomy;
• The requirement to reschedule appointments.

ADRs were identified when interpreting blood test results, as reported by the hospital laboratory, and symptoms described during telephone consultations. Changes to the frequency of blood testing, dose reductions, and/or discontinuation of treatment were recorded, including action taken by the pharmacist and reported to the gastroenterologists.

Patient questionnaire

A patient satisfaction questionnaire comprised of 12 questions was designed, piloted and validated by the University of Bradford. When the study was designed in 2014, a paper questionnaire was considered the most appropriate pathway to obtain the required data.

The aim of the questionnaire was to explore patients’ perception of the accessibility, convenience and overall satisfaction of the clinic using a 5-point Likert scale, where 1 was ‘very poor’ and 5 was ‘very good’, and their preferred method of communication (e.g. telephone, voicemail, SMS, e-mail, face-to-face and letter). The printed questionnaires were posted by mail with a pre-paid envelope to all participants, who were asked to keep them anonymous. Questionnaires were therefore not coded and non-responders were not re-addressed, to avoid data duplication.

Patients’ perception was also measured through a follow-up questionnaire in 2019 via an SMS using NHSmail in SurveyMonkey. The number of questions was reduced from 12 to 4, to reduce the time required to complete the questionnaire, and they were worded the same as those in the original.

Results

A total of 81 patients; 36 men and 45 women, were included in the 2014 study and had a median age and weight of 46 years and 81.3kg, respectively. The distribution of patients comprised 58% (n=47) with CD, 34.6% (n=28) with UC and 7.4% (n=6) with AIH. The study showed 81.5% (n=66) of patients were thiopurine naïve, where 18.5% (n=15) required dose optimisation.

Blood test analysis

The total number of routine blood samples (comprising full blood count, urea and electrolytes and liver function tests [LFTs]) collected was 304. In total, 60.5% (n=49) of patients reached the therapeutic target without biochemical or clinical complication. However, 32 patients (39.5%) failed to reach the therapeutic target during the study period because of side effects or intolerance to the medication. Of the patients presenting issues with the medication, 9.9% (n=8) developed hepatotoxicity and 29.6% (n=24) were thiopurine-intolerant or did not respond clinically. The thioguanine nucleotide analysis on patients who were thiopurine-intolerant or did not respond clinically showed 13.6% (n=11) were sub-therapeutic, 14.8% (n=12) supra-therapeutic and 1.2% (n=1) did not adhere to treatment (see Table 1 below).

Attendance for phlebotomy

Phlebotomy service was offered to the patient during their first face-to-face visit and was arranged in hospital or with their GP at a convenient time for the patient, but before the follow-up virtual clinic. Overall, 260 tests (85.5%) were attended as scheduled, with the remainder requiring rescheduling. After rescheduling, a further 39 blood tests (14.5%) were completed.

Adverse drug reactions

Neutropenia (neutrophils <1.7 x 109/L blood) and leukopoenia (white cells <3.7 x 109/L blood) were observed in 6.9% (n=21) of blood tests. Results were interpreted and actions taken to re-address, mainly through dose adjustment. Discontinuation of thiopurines was required in 2.5% (n=2) of patients owing to significant and persistent myelosuppression. Transaminitis developed in 9.9% (n=8) of patients, requiring temporary cessation of treatment and subsequent reintroduction of thiopurines (at 25% of the target dose) combined with allopurinol. There were no cases of hepatotoxicity requiring discontinuation of thiopurines when combination therapy was used.

Patient questionnaire in 2014

Patient questionnaires were posted to all study participants (n=81) in a pre-paid envelope. The response rate was 51.9% (n=42).

For the question “What is your preferred choice of communication?”, results showed that overall, 52.4% (n=22) of the respondents preferred telephone clinics, contrasting with those who favoured SMS (26.2% [n=11]), email (9.5% [n=4]), conventional face-to-face visits (7.1% [n=3]), written documentation (2.4% [n=1]) and voicemail (2.4% [n=1]).

Respondents scored their overall satisfaction of the clinics: excellent 73.8% (n=31), good 21.4% (n=9), average 4.8% (n=2), poor 0% and very poor 0%.

Re-evaluation patient questionnaire in 2019

A total of 373 patients were included in the re-evaluation. The response rate was 35.9% (n=134). Overall, 88% (n=118) of the patients found the service easily or very easily accessible and 82% (n=109) found it very helpful or extremely helpful (see Tables 2 & 3) .

Responders were asked to rank preference of communication between seven options and the majority preferred a virtual clinic with telephone calls, emails and SMS over the traditional face-to-face method (see Table 4 [results analysed from 133 responses; 1 person skipped the question]).

In the final question, patients were asked to rate their overall satisfaction with the virtual clinic. This rating was high, with an average score of 4.46/5.

Finally, the survey included a free-text option to provide direct feedback. Some patients’ additional opinion confirmed their satisfaction with the service:

• “Excellent work”;
• “Very friendly and helpful pharmacist”;
• “The service is an asset to the hospital”;
• “Great support & would like to see this service continue as a touchpoint between appointments”;
• “Superb service, I cannot fault them”;
• “I have always found the service excellent, the pharmacist has been extremely helpful in explaining my bloods and medication … to a non-medical professional, he always clearly explains the process, the plan and next steps, I really appreciate it.”

Discussion

Immunomodulator therapy for the treatment of IBD and AIH offers improved symptom control and quality of life, reduced in-patient admissions, bowel surgery, progression of underlying disease, and exposure to complications associated with long-term steroid use (e.g. increased risk of infections or osteoporosis)^[5]
. The typical environment for the monitoring of such therapies is in a face-to-face outpatient IBD nurse-led clinic.

The integration of the clinical pharmacist within the gastroenterology team at Barnsley Hospital NHS Trust in 2012 has enabled the safe induction of thiopurine therapy evidenced by our study results.

Since our initial study in 2014, the IBD clinic has integrated the monitoring of other immunosuppressants (e.g. methotrexate, mycophenolate and tofacitinib). In 2018, there was a renewed focus to take a holistic view of current IBD treatment and identify patients whose response to therapy could be sub-optimal ahead of attending follow-up in the outpatient setting. A remote service has the potential to improve access for patients to specialist NHS services and offers prompt clinical support. Therefore, our remote service has continued throughout the recent COVID-19 pandemic without interruption.

This study challenged the notion that therapeutic drug monitoring can only be performed in a traditional consultant-led or specialist nurse gastroenterology clinic. Results show that pharmacists can lead immunomodulator clinics, providing there is appropriate consultant guidance and approved treatment pathways.

Blood monitoring

Excellent compliance was demonstrated with the blood monitoring schedule. Of the 304 routine blood tests performed, 85.5% (n=260) were completed at the scheduled appointment and an additional 12.8% (n=39) were completed at a rescheduled appointment. Reasons for the missed appointment included the patient forgetting to arrange the blood test, there being no time to attend phlebotomy or time to call a GP to arrange as scheduled. Only 1.7% (n=5) could not be rescheduled owing to non-attendance. Thiopurines were discontinued in those patients who failed to attend blood monitoring appointments. Compliance associated with therapeutic monitoring in our study is superior than in a nurse-led IBD thiopurine monitoring service study that had a reminder rate of 43% (350 out 810 total blood tests)^[20]
.

TGN analysis

To the author’s knowledge, this is the first study that explores pharmacist-led TGN analysis in a dedicated thiopurine drug-monitoring service. The therapeutic target was not achieved in 39.5% (n=32) of patients owing to intolerance, side effects, or lack of clinical response. TGN analysis allowed further investigation of this sub-group of patients, in which therapeutic target was not achieved. Sub- or supra-therapeutic TGN levels prompted tailored thiopurine dose optimisation^[21]
. TGN analysis also identified one patient who was non-compliant with treatment. This simple observation has clinical relevance because it highlights any discussion about more aggressive or costly forms of immunosuppression.

As part of the service, results were discussed over the phone and no extra consultations were arranged at any time in medical outpatients. If the medication regimen required adjusting or further discussion, a brief consultation was agreed with the patient when collecting medication from the outpatient pharmacy.

Adverse drug reactions

ADRs associated with thiopurines were principally dose-related as variation in thiopurine induction dosing exists. Some healthcare professionals prefer to gently titrate the dose of thiopurines following serial clinical and biochemical review, while others prefer to start their patients at the target dose. The protocol for this study was agnostic to either strategy^[6]
. 

The prevalence of hepatotoxicity at standard thiopurine doses was found to be 9.9% (n=8), consistent with a nurse-led thiopurine monitoring study, which found similar results^[20]
. In our study, thiopurines were discontinued until LFTs had normalised in those patients who developed hepatoxicity. Dual therapy with allopurinol (and 25% of the standard thiopurine dose) successfully allowed these patients to continue treatment without further transaminitis^[8]
.

Myelosuppression was observed in 6.9% (n=21) of patient blood tests. However, most abnormalities were mild and transient, with only 2.5% (n=2) of patients experiencing significant myelosuppression that required treatment cessation and referral to the relevant gastroenterology consultant. These examples highlight how pharmacists can identify myelosuppression at an early stage and advise on treatment decisions to ensure the highest standards of patient care.

Patient satisfaction questionnaires

These were designed and validated by the University of Bradford, following review of existing validated IBD questionnaires developed to evaluate patient satisfaction with healthcare services (e.g. CACHE or QUOTE-IBD)^{[22],[23],[24]}
. Patient satisfaction is increasingly seen as an important marker of the quality of clinical care. A patient who is satisfied with their treatment is more likely to adhere to treatment; therefore, achieving better clinical outcomes^[25]
. This is particularly pertinent in those patients with chronic and life-long diseases, such as IBD.

The patient questionnaire response rate was 51.9% (n=42). This satisfies the minimum required response of 50% for social research postal surveys^[26],[27]
. The results confirmed that the majority of patients (88.1%; n=37) preferred the more modern approach to communication, using media such as SMS 26.2% (n=11), email 9.5% (n=4) and remote telephone consultations 52.4% (n=22). To put this into perspective, 7% of respondents preferred the conventional face-to-face outpatient approach. There are many reasons why this may be the case. Younger patient populations may have work and family commitments that preclude regular visits to the hospital. Furthermore, the ubiquitous use of smartphones, email and cloud services means that remote monitoring can be readily embraced. Nevertheless, this study was unique in offering patients a choice about how they should engage and communicate with the pharmacist, a mechanism that encouraged further participation in the project.

Overall satisfaction of the clinic was deemed ‘excellent’ by 73.8% (n=31), and ‘good’ by 21.4% (n=9) of patients. This gives credibility to the idea that giving patients a choice and a dialogue in the management of their condition leads to satisfaction, and that virtual thiopurine clinics can play an important role in the delivery of quality care.

In 2019, we re-evaluated the service and sought the feedback of all IBD patients treated at Barnsley Hospital NHS Foundation Trust via an electronic survey.

A total of 373 patients were invited to participate and the response rate was 35.9% (n=134). Overall, 88% (n=118) found the service easily or very easily accessible, and 82% (n=109) of patients found it very helpful or extremely helpful. Responders were asked to rank preference of communication between seven options and the majority preferred a virtual clinic with telephone calls, emails and SMS over the traditional face-to-face method. Finally, the overall satisfaction rate with the clinic was high, with an average score of 4.46/5. Finally, many patients supported the idea of using flexible approaches in supporting their therapy, showing the model can be replicated and adapted to other specialities. 

Limitations

The initial study period was three months (May, June and July 2014) and, as patients usually present with myelosuppression when treatment is initiated, this time period was acceptable to meet the study aim^[6]
. The number of participants recruited (n=81) was not large; however, the number of blood tests (n=304) obtained during the evaluation period met the aims. With regard to the data, another limitation of the study is the period of time since collection of the initial data in 2014 to the date of publication, explained by work pressures and the lead author taking a career break. However, as discussed earlier, the data presented is still relevant to today’s practice.

The study included the results of two patient satisfaction questionnaires conducted in 2014 with 81 patients with 42 respondents, and a second in 2019 with 373 patients and 134 respondents. Reminder letters were not posted to participants in 2014, so this number could have been affected by factors such as patients forgetting to complete the survey or losing the paper at home. The 2019 electronic questionnaire and reminder text was sent to all participants to minimise these issues. All IBD patients registered with a mobile phone in the IBD database were contacted. Nevertheless, the electronic questionnaire was designed to be opened only with a smartphone, which limited our response rate, as we were not aware if patients had a smartphone to access the survey.

Conclusion

The introduction of a dedicated pharmacist-led virtual thiopurine clinic has optimised clinical outcomes, treatment adherence and monitoring, patient safety and education, while reducing the demand on overstretched resources in the outpatient setting. Pharmacist-led clinics are safe and cost-effective and reduce the number of face-to-face consultant or specialist nurse outpatient appointments.

Early detection of myelosuppression and hepatoxicity is a critical function of the clinic, and rapid treatment decisions, thiopurine dose adjustments, and deferred introduction of combination therapy with allopurinol can be made under the pharmacist-led care model. Local IBD services have also benefited, as the pharmacist monitors drug response and advises on drug adjustments according to clinical response. In the meantime, specialist nurses can follow up more patients via telephone helpline or through the outpatient clinic, freeing up more clinic slots for consultants to see more patients. Overall, the optimisation of therapy in patients with IBD and AIH may also prevent costly treatment escalation over the longer term. These benefits have been reflected in patient satisfaction scores.

Future development of the nurse-pharmacist partnership should be studied to explore how the collaboration of the specialist pharmacist and nurse working together can enhance NHS services and improve patient care.