AUTH/3568/10/21 - Voluntary admission by Roche

Typographical error regarding dosing within the Gazyvaro leavepiece

  • Received
    06 October 2021
  • Case number
  • Applicable Code year
  • Completed
    07 December 2021
  • Breach Clause(s)
  • Sanctions applied
    Undertaking received
  • Additional sanctions
  • Appeal
    No appeal

Case Summary

Roche made a voluntary admission in relation to a typographical dosing error in a Gazyvaro (obinutuzumab) leavepiece (ref M-GB-00000509) which the company submitted was for nurses.

Gazyvaro (obinutuzumab) was a Type II humanised anti-CD20 monoclonal antibody indicated for the treatment of certain patients with follicular lymphoma (FL) or chronic lymphocytic leukaemia (CLL). One vial of 40mL concentrate contained 1,000mg obinutuzumab, corresponding to a concentration before dilution of 25mg/mL.

Roche stated that the incorrect information about the Gazyvaro dose within the leavepiece at issue was in a chart on page 6 which set out the infusion rates for CLL patients. The chart gave information on the dose, Gazyvaro volume and final infusion volume for cycle 1, day 1, day 2, day 8 and day 15. It also had information for cycles 2-6. The front page of the leavepiece referred to the product by name, Gazyvaro, followed by ‘checklist for the treatment of patients with follicular lymphoma (FL) & chronic lymphocytic leukaemia (CLL)’. The Gazyvaro dose for cycle 1 day 1 on page 6 was given as 1000mg instead of 100mg.

Roche stated that the error was identified by an employee on 1 September 2021. Roche submitted that all other values including the Gazyvaro volume, final infusion volume and the infusion rates were correct as were any other references to dosage contained within the leavepiece.

Roche stated that it was committed to the appropriate use of medicines, protecting the safety of patients and strove to maintain high standards in the ethical promotion of its medicines. As soon as this single typographical error was identified the leavepiece was immediately withdrawn from use and a full investigation initiated. Details were provided.

Roche submitted that Gazyvaro was launched onto the UK market in June 2015. The material in question was therefore not part of the launch materials which would have introduced the health professionals to the product, and inter alia, its dosing regimens. By 2019 the product could therefore be considered as more established on the UK market and its use and administration more familiar to UK hospital practices.

Roche stated that for a product such as Gazyvaro to be administered in hospital, well established and appropriate prescribing, procurement, preparation and administration procedures had to be in place. These procedures would be developed in line with the relevant data and safety information. Manufacturers’ promotional items would have been most commonly utilised as an aide memoire for discussions with Roche representatives on infusion rates. It was therefore felt that the item in question would be sufficiently far enough removed to have a direct impact on well-established practices and more importantly the administration of an incorrect dose. Were this item to be used to guide in procedural development, it would have been used in conjunction with other documentation such as the summary of product characteristics, hospital protocols, and relevant data sheets; the combination of which would have readily highlighted the typographical error.

Roche submitted that Gazyvaro was typically reconstituted and prepared by pharmacy or was outsourced to a compounding/aseptic unit. The primary audience for page 6 was a haematology nurse who would be involved in the administration of the product once reconstituted rather than the writing of a prescription or preparation of the product. As such the primary intent of page 6 was to detail the infusion rates that those administering the product would need to use to obtain the correct dose, rather than focus on the dose prescribed.

Roche stated that it had been considered that were a nurse to use this guide to draw up a dose, the volume of 4mL as per the volume stated in the column headed ‘Gazyvaro volume (mL)’ would result in the correct dose of 100mg rather than 1000mg given; Gazyvaro was a 25mg/mL solution. It was also noted that if this item were to be used to draw up a dose, the discrepancy between volume and dose would thus highlight the typographical error.

Upon review of the above and the wider investigation and following withdrawal and the permanent discontinuation of the material, it was deemed appropriate to notify the relevant health professionals who might have received the material containing the single typographical error and recommend its destruction. As such a communication in the form of a written letter identifying the material in question and outlining the single typographical error was sent to each unit in the UK with the potential to administer Gazyvaro.
In light of the above Roche considered this a breach of the Code on the basis that the information provided was inaccurate due to the single isolated typographical error.

Whilst it was identified that inadvertent human error had likely led to the single typographical error remaining present in the material at certification, Roche considered that the high standards expected of the company had not been maintained in breach of the Code.

Roche submitted that whilst it was regrettable to find a single isolated typographical error in a piece of material, it was confident following an exhaustive safety investigation that there was no direct impact to patient safety, and felt the immediate and robust actions taken through comprehensive investigation with corrective measures highlighted the high standards Roche constantly strove to maintain. As such Roche believed the high standards expected and confidence in the industry had been upheld.

The detailed response from Roche is given below.

The Panel noted that the leavepiece at issue was certified and approved for use on 22 July 2020. Roche submitted that the error was identified on 1 September 2021 as part of the printing of a new batch of leavepieces. Roche had instigated a withdrawal but as the existing stock had been distributed there was no stock returned by representatives. Roche explained that health professionals were informed of the error on 6 October 2021 by a letter sent by post to haematology units in the UK which administered Gazyvaro. The units were selected by the account list covered by Roche representatives and those units purchasing the product.

The 14 page leavepiece first detailed dosing information for FL. Page 5 of the leavepiece set out the dosing and schedule for CLL patients. Page 5 was clear that for cycle 1 the first 1000mg was split over two infusion bags, with 100mg on day 1 followed by 900mg on day 1 if there were no infusion related symptoms. The 900mg dose would be given on day 2 if there were infusion related symptoms. Similar information was given about the rest of cycle 1 (with 1000mg on day 8 and day 15) and Day 1 of cycles 2-6. Page 5 gave no information about the Gazyvaro volume, the final infusion volume or infusion rates; it only referred to infusion bags.

The Panel noted the error on page 6 where the first dose of Gazyvaro was listed as 1000mg instead of 100mg. Roche submitted that it was an isolated error. The Panel noted Roche’s submission regarding the accuracy of the other information on the page and the protocols and arrangements in hospitals regarding the preparation of Gazyvaro by the hospital pharmacist prior to its administration. It also noted Roche’s submission that page 6 of the leavepiece was clear about the Gazyvaro volume for cycle 1, day 1; health professionals using the information about the Gazyvaro volume (4ml) would mean that the correct dose of 100mg would be used rather than 1000mg, given Gazyvaro was a 25mg/ml solution.

Pages 7 and 8 of the leavepiece were headed ‘Preparing your FL and CLL Patient for their treatment’ with page 8 including an additional section for CLL patients. This referred to the need to prepare CLL patients for the possibility of infusion related reactions (IRRs), particularly in advance of the first infusion on day 1 (and day 2 if infusion was split over 2 days) and the need to monitor the patient closely especially for the first 2 hours. The leavepiece also included information about premedication and managing IRRs.

The Panel was very concerned about the error on page 6 and considered that it could have led to patient safety issues, particularly as the error related to the first dose of the medicine. Page 6 was inaccurate and misleading. The Panel therefore ruled breaches of the Code including that high standards had not been maintained as acknowledged by Roche.

The Panel noted the actions taken by Roche once the error was discovered. These included withdrawing the leavepiece, informing health professionals of the error and checking the Roche safety database with regard to the impact of the error. Roche submitted that there was no evidence that the error in the leavepiece had resulted in an increase in reports of errors of administration in patients. The Panel noted that Clause 2 was reserved for use as a sign of particular censure and that the supplementary information to that clause listed prejudicing patient safety as an example of an activity likely to be in breach of Clause 2. The Panel noted that there was more than a year between certification of the material and identification of the error and that during that time more than five hundred hard copies of the leavepiece had been distributed to health professionals as well as electronic versions of the material downloaded from Roche’s website. The Panel queried why it took Roche more than one month between identification of the error in the material and notification to health professionals of the error. The Panel noted its comments and rulings above and considered that the dosing error in the material might have prejudiced patient safety and therefore a breach of Clause 2 was ruled.