[Table of Contents]

Volume: 27S1 • July 2001

Report of the Xenotransplantation Surveillance Workshop
Infection Control Database and Sample Archiving

II. SURVEY RESULTS AND DISCUSSION

The main goal of the workshop was to capture the views of the various participants on certain issues and not to mould opinion or reach any kind of consensus. Generally, most issues elicited a wide range of views and comments. These responses are instructive because they help identify where problems lie and what issues need to be addressed in future deliberations and, presumably, in guidance documents.

Patient Registry and Adverse Event Reporting Database

Participants noted that the design of a xenotransplantation surveillance database will depend greatly on what purpose(s) it is to serve. Surprisingly, some participants considered enhanced surveillance measures by HC to be a research project on safety assessment. However, since xenotransplantation carries an as yet undefined risk to the public in the way of a new epidemic, the assessment and containment of this risk at the clinical trial level is essential for the health protection of Canadians. Carefully designed and closely monitored clinical trials are viewed as the only means to adequately assess the risk of xenotransplantation procedures to public health and safety⁽¹⁾. Furthermore, clinical trials should not be allowed to go ahead without public consultation, oversight by a national body, and precautionary measures to assess and contain the risk of infectious diseases to third parties⁽¹⁾.

One could envision the initial registry of the small number of recipients being merely paper-based; however, once several trials are approved and under way, it will become necessary to develop an electronic version capable of linking various data sets. Discussion on the registry at the workshop identified two functional types of data elements, those contained within the national registry and those linkable or accessible through common identifiers to other databases, such as an archiving database or the medical records of the recipient at the local site. Most participants felt there would not be a need to have the full medical history or patient file at the national registry, since the relevant information could be ascertained upon inspection, investigation, or linkage. It was felt that if the registry database became too detailed or complex, the ability might be lost to respond to an early sign or trend indicating transmission of an infectious agent and that the costs and potential for information overload would be greater.

Most respondents indicated that the demographics of the source animals and tissues should be accessible or linkable, but not necessarily held in the national registry. On the other hand, there was general agreement that the demographics of the recipient (age, sex, hospital site of xenograft, type of xenograft) should be included. The Proposed Canadian Standard for Xenotransplantation calls for the inclusion of the following data elements: xenotransplant program identification number, recipient coded identification, race, sex, age and date of birth, SIN number, ABO blood grouping, organs received and cause of organ failure, date and time of transplantation, unique source animal identification number and all data relevant to source animal parentage, husbandry, and testing for transmissible (or potential) pathogens⁽⁸⁾. Although the proposed standard recommends that source animal information be contained within the national registry, many participants thought that linkability or accessibility only was needed. They did not feel that other data elements, such as whether the patient was immunosuppressed, type of immunosuppression used, the patient's transfusion or transplantation history, and history of human infectious diseases or cancer, were important enough at this time to be in the national registry.

All responding participants indicated the critical importance of recording serious adverse events in the registry, and most felt that other AEs should also be included. However, the types of AEs to be submitted to the patient registry will require further clarification and definition (see also below).

In conclusion, participants agreed with the minimal data elements to be submitted to the national registry as outlined in the Proposed Canadian Standard for Xenotransplantation⁽⁸⁾, with the potential exception of the xenograft source demographics.

Definition of Adverse Event for Xenotransplantation

The definition of an AE received much debate. In the Proposed Canadian Standard for Xenotranplantation an AE is defined as "an undesirable outcome directly or indirectly related to the allo/xeno graft (e.g. infection, disease transmission, graft failure)"⁽⁸⁾. It is assumed that death, though not listed, would be considered "undesirable". As an alternative, the following definition was suggested in the survey (see the enclosed survey template in Appendix B): "any notable change in the recipient's health or well-being, reversible or not, requiring medical attention or intervention, inclusive of death, disability, disease transmission, infection or hospitalization whether directly or indirectly related to the xenograft protocol". Many participants chose one of the two similar definitions provided but listed caveats or additions, such as graft failure or the diagnosis of cancer.

One suggestion during the workshop discussion was to adopt the WHO or other internationally agreed upon definition; however, at present the WHO has not released its definition of "adverse event", although international discussions are under way. Some participants preferred that the sponsor/clinicians define for themselves what is reportable or decide whether the AE was related to xenotransplantation before reporting. For AE reporting in the enhanced surveillance of vaccines, there is a separate panel of experts that reviews the AEs reported to decide whether they are related to the vaccination process⁽¹²⁾. If a serious AE is rare or unusual, this decision may be very difficult. Some workshop participants suggested that all serious AEs should be reported, closely monitored, and/or investigated, since xenotransplantation is new, and it may be difficult to recognize when the AE relates to the procedure and/or xenozoonosis. Some felt that any abnormal laboratory test results or any persistent AE, such as fever, should be reported, whereas others cautioned against flooding the system. The challenge is to ascertain the true AE related to xenotransplantation as opposed to background noise. Some participants also felt that if family members suddenly experience an unexplained illness this should also be reported.

The UKXIRA guidelines had defined "untoward event" as a suspected xenozoonosis "and" confirmed xenozoonosis⁽¹⁴⁾. The UKXIRA has recently changed this to "and/or", as most xenozoonosis would not be confirmed before an investigation is conducted. However, restricting AE reporting to xenozoonoses may be counterproductive, as there are no uniform clinical criteria for suspecting a xenozoonosis, especially with regard to unknown or latent viruses. There is no easy way to discriminate human from zoonotic infections based on the clinical picture. Under these circumstances, either all AEs or, what is more likely, few AEs would be reported and investigated. Without further clarification, the UK definition of an adverse event seems inadequate, as even deaths may go unreported. Certainly, analysis of autopsy tissues from xenograft recipients would provide essential sources of information about transmission of endogenous and endemic pig viruses to humans and would be key to the assessment of the infectious disease risks of xenotransplantation. At the very least the UKXIRA would need to reconsider whether deaths should or should not be reported and infectious agents investigated on autopsy samples, regardless of causality.

For AE reporting, the Proposed Canadian Standard for Xenotransplantation states that "the following adverse events shall be immediately reported (i.e. reporting is mandatory and immediate): potential or confirmed xenozoonosis in the recipient or positive test for infectious agent(s) in the source animal; appearance in the recipient of a reportable infection such as HIV, hepatitis B or C, tuberculosis, or any other infection(s) including zoonoses; appearance in the recipient of a new cancer (excluding basal cell or squamous cell skin carcinomas); primary non function of the transplanted xeno-organ; and death of the recipient"⁽⁸⁾. The definition of adverse event for xenotransplantation and the mandatory reporting differ considerably from the Adverse Drug Reaction definition for INDs, in which the reporting of serious AEs within clinical trials contains the element of "unexpected" and suggests that "a causal relationship is at least a reasonable possibility"⁽⁵⁾. For medical devices, the report is mandatory when it relates to the failure of the device or its effectiveness, and has led to the death or serious deterioration in the state of health of a patient⁽⁷⁾. Clearly, the scope, reporting, and testing requirements relating to AEs in xenotransplantation protocols need further definition and clarification in a guidance document before xenotransplantation clinical trials are considered in Canada. Given the likelihood that most pig herds are positive for PERVs (blood and/or tissues) and probably other endemic viruses, such as pCMV, it remains to be decided whether the presence of these viruses (in the source animal or the herd) should or should not preclude xenotransplantation.

Linking Adverse Events to Xenozoonosis

Most workshop participants thought that death, cancer, autoimmunity, disabilities such as paralysis, graft rejection, rashes, and fevers might be causally linked to an infectious agent (human or porcine). The notion that viruses may be implicated in a number of autoimmune conditions and chronic inflammatory diseases is not new. Several human endogenous and exogenous retroviruses have now been implicated, including a new exogenous virus called human retrovirus five (HRV-5), which is implicated in arthritis and was recently described by Weiss et al.⁽¹⁸⁾. The concern has been raised before about PERV pseudotyping by other human retroviruses present in the host and the subsequent release of replication-competent PERVs. Furthermore, retroviruses can be oncogenic (cancer-causing). Patients with autoimmune, chronic inflammatory processes or cancer may need to be excluded from or (if included) more closely monitored in xenotransplantation clinical trials until such time as PERVs can be excluded from source herds and xenografts.

Other AEs potentially linked to xenozoonosis include a range of illnesses and syndromes (in recipients and close contacts) and could include any abnormal laboratory value, particularly if there is a positive test result for pig infectious agents (specific or nonspecific). One participant went so far as to say that any AE might be causally linked to an infectious agent. The Proposed Canadian Standard requires mandatory creation of Standard Operating Procedures (SOPs) to be used for the local investigation and follow-up of suspected infections and communicable diseases⁽⁸⁾. If local xenotransplant teams are free to develop their own standards, a non-uniform set of reporting and investigative processes will likely result. A national investigative team may be better suited to perform the investigation and determine, in an unbiased way, whether a porcine infectious agent is involved or is potentially linked to the AE. Analysis in an aggregate fashion by a national authority may improve consistency and facilitate the earlier detection of trends. Nevertheless, it is expected that there will be general difficulty in confirming whether an adverse event relates to an unknown xenozoonosis. Accordingly, there seems to be little value in limiting the definition of an adverse event to a xenozoonosis.

Criteria for Follow-up Investigation of Potential Zoonosis

Essentially any unexplained AE could be followed up for a potential zoonotic agent. Some participants thought that only if there is some evidence for a zoonotic infection (such as that established through screening methods) should there be follow-up. Several participants believed the sponsor or local expert team of investigators should decide. In keeping with the recommendations of the National Forum⁽¹⁾, many participants felt that when a death occurs, an investigation for pig infectious agents should be performed on autopsy samples. Some suggested that a diagnosis of cancer or auto-immunity may also warrant such an investigation. Thus, the criteria for a follow-up investigation will need further debate and resolution. As well, efforts should be made to harmonize these internationally⁽¹⁾. Whatever criteria are developed, they should be established in advance of clinical trials and made available through an interim policy released by HC on the conduct of clinical trials involving xenotransplantation. Although it may be relatively straightforward to investigate known agents with defined sequences, unknown pig infectious agents may prove more difficult to confirm.

Time Lines for Reporting Data and Adverse Events to Registry

In general, there was a consensus that serious adverse events should be reported immediately, in agreement with the Proposed Canadian Standard for Xenotransplantation⁽⁸⁾, but the definition of "immediately" was somewhat murky. During the open discussion most participants agreed that 24 hours should be the definition of "immediately", whereas in the survey report there was a range from 12 hours upon receiving confirmation to 72 hours. With regard to reporting by 24 hours, one participant questioned whether anyone at HC would be available on the weekends. If the criteria for follow-up investigation are clarified in the interim guidance document and it is stipulated what enhanced surveillance is mandatory (including required samples, where to send the samples, when a follow-up investigation is needed), then perhaps a 72-hour maximal time line would be acceptable. Again, whatever minimal requirements are needed, these should be established before the authorization of clinical trials in Canada.

There was a lot of variability in participants' responses about serious versus not-so-serious AEs. Many participants felt that quarterly instead of annual reporting should be required for xenotransplantation clinical trials, because some less serious AEs, such as a flu-like condition, when analyzed in an aggregate fashion, may provide the earliest indication of a xenozoonosis epidemic.

Sample Archiving

Many participants preferred the word "archiving " in the survey questionnaire to "monitoring", since they felt that routine or active testing by a national body was not warranted. As well, many felt that there is no need to duplicate archiving at a national and local site. Instead, samples could be made available in the event of an investigation, or duplicate samples for a national body could be held locally. The latter is consistent with the new U.S. PHS guideline but not with the recommendations of the National Forum on Xenotransplantation⁽¹⁾, which greatly favours separate retention of locally and nationally archived duplicate samples. The Proposed Canadian Standard for Xenotransplantation⁽⁸⁾ states that "in the absence of a central facility, designated public health biologic specimens (serum, plasma, leukocytes, and tissues of the recipient, if available) should be archived with appropriate safeguards to ensure long-term storage and an efficient document system for the prompt retrieval and linkage of data to medical records of recipients and source animals."

For recipient samples, most participants agreed that baseline, routine, and investigative samples would need to be archived on all patients and that serum (5 aliquots of 0.5 mL for both the local/national sites), plasma (same quantity as for serum), and peripheral blood mononuclear cells (PBMC, at least 3 aliquots of 1 million cells for local and national investigators) should be stored. Other suggestions for archiving included biopsy and autopsy specimens if appropriate/available, and RNA and DNA samples; few felt that urine samples would be useful. Some suggested that there is no need for a national archive, whereas others maintained that there should there be a national archive and all testing be done by HC. The ability to obtain blood should be considered, and the volumes collected need to be practical. Indeed, one participant suggested that the sponsor should determine amounts of samples to be collected. It was also proposed that expert committees should determine which tests should be performed under what circumstances, since the total sample volume would be limited, and certain samples may be critical for the investigation. Since not all tests would be available in a validated and standardized format, it may be preferable that an independent approval committee decide what tests are to be performed.

Like the Proposed Canadian Standard, the UKXIRA guidelines propose lifelong monitoring. If the storage of samples is recommended for at least a minimum of 50 years, as per the new U.S. PHS guideline⁽¹⁰⁾, the question arises of whether age restrictions on xenograft recipients should be required (favouring elderly patients), or whether initial trials should be initially limited to terminal patients. As well, a mechanism is needed to ensure the availability of specimens for this length of time, probably through storage at a national archive site.

In terms of the source animal samples to be archived, many respondents felt the samples should be held at the local site or by the sponsor, and HC be allowed access in the event of an investigation. Assuming the animal is sacrificed at the time of cell, tissue, or organ harvest, as was recommended at the National Forum⁽¹⁾, then tissues such as lymph node, spleen, bone marrow, and any other tissue relevant to the type of xenograft used could also be archived, along with plasma (5 aliquots of 0.5 mL), serum (5 aliquots of 0.5 mL) and PBMC (5 aliquots of 10 million cells). Because of the concern that hepatitis E is endemic in herds and transmissible to humans, some participants suggested collecting fecal samples from the source animal for PCR tests, since this is the most sensitive and reliable method for early detection⁽¹⁷⁾.

For close contacts, baseline and investigative samples should be taken (as with recipients), but only a few respondents felt that routine annual samples were necessary. Most suggested that local storage of close contact materials would be sufficient.

Sample archiving for health care workers was judged to be similar to that for close contacts, with additional samples taken in the event of needlestick injury. Many thought those at highest risk might be monitored (archived and/or tested) more frequently. The U.K. guidelines suggest that, aside from baseline blood samples, samples on health care workers should be archived and tested only if there is significant exposure to bodily fluids of xenograft recipients (such as for surgeons) or if there is any untoward event⁽¹⁴⁾.

Schedule for Recipient Archiving/Testing

Aside from the need for a baseline sample on all recipients and for investigative samples for an AE, the Proposed Canadian Standard for Xenotransplantation⁽⁸⁾ does not detail a minimally acceptable schedule for the archiving or testing of recipient samples. Workshop respondents were given a choice of two schedule types for sampling patients, or they could develop their own schedule (see attached survey template in Appendix B). In the initial trials, the schedule is expected to be more intense in order to collect rudimentary information as to which samples are the most informative. Once the initial clinical risk assessment is achieved, it may not be necessary to collect samples as frequently in the immediate post-xenograft period. Many participants agreed on weekly or biweekly sampling in the first 3 months, then monthly for the next 3 to 9 months, and then annually. Most felt that at least a sample at 1 month and at least twice in the first year would be necessary.

The U.K. guidelines recommend archiving only for baseline, month 1, month 6, and then annually; this is based on the 1996 U.S. PHS guideline⁽⁹⁾. On the other hand, the U.K. guidelines call for testing of samples at 0, 1-2 days, 2, 4, and 6 weeks, 3 and 6 months, and annually thereafter. The rationale for testing but not archiving of some samples is unclear in the UKXIRA guidelines. The revised 2000 U.S. PHS guideline⁽¹⁰⁾ (in section 4.1.2) now states that for archiving there should be two separate baseline samples (preferably separated by a month and where one sample is time 0), and that samples in the immediate post-transplant period should be archived along with those from 1 and 6 months, 12 months, 24 months, 60 months, 120 months etc.

With regard to testing, the revised PHS guideline recommends that if xenogeneic infectious agents are known or suspected to be present in the xenotransplantation product, then active screening of the recipient should be performed frequently in the immediate post-xenotransplantation period. They cite, as examples, the 2, 4, and 6 week specimens and suggest screening could include endogenous retroviruses, herpesviruses, and papillomaviruses. Since all pig herds harbour PERVs and many, if not all, harbour endemic viruses of one kind or another⁽¹⁷⁾, then as a standard (for the U.S., U.K., and probably Canada), archiving and active testing for endogenous and endemic pig infectious agents may need to be performed minimally at baseline, 2 weeks, 4 weeks and 6 weeks, 6 months, annually (for 2 years), and then every 5 years. This is one possible interpretation of the revised U.S. guideline.

The word "monitoring" in our survey template was meant to include testing and archiving. However, many participants felt that "monitoring" should be changed to "archiving", and this would eliminate the need for active testing. However, as mentioned, the U.S. guideline (referred to by the UKXIRA⁽¹⁴⁾) makes it very clear
that active testing and archiving are now recommended as part of the monitoring. This is particularly important in the immediate post-xenograft period, and especially since most herds probably harbour known endogenous^(19-25,31) and endemic^(16,17) viruses.

Most participants felt that close contacts need baseline and annual sample archiving, and testing and archiving in the investigation of an AE. Some, however, thought quarterly or monthly sampling for the first 3 months could be considered, presumably to allow for the initial assessment of risk of sexual or household transmission. The U.K. guidelines recommend that a baseline and a sample at 1 year should be archived; testing and further investigational archiving would be necessary only in the event of an AE potentially related to xenozoonosis⁽¹⁴⁾. The U.S. PHS guideline suggests that only a baseline sample is required on health care workers, and testing and further archiving is only necessary in the case of an AE⁽¹⁰⁾. The U.S. PHS guideline does not cover close contacts.

Pre-Clinical Work-Up on the Source Herd

When referring to the xenograft animal materials or animals, most participants preferred the use of the term "source" to "donor". Many felt that a sentinel health surveillance program is required, possibly with some sentinels being temporarily immunosuppressed (i.e. for up to 2 months) and observed; others did not think this was feasible. The Proposed Canadian Standard for Xenotransplantation suggests, among other measures to isolate the source herd from infectious agents, that a sentinel surveillance program should be encouraged as a standard⁽⁸⁾. Some participants proposed that the sentinel program be instigated on the biosecure herd and not on each litter, since conclusions cannot be drawn for about 15 years (the life span of a pig), and it would be best to have this assurance of safety long before, or as soon as possible after, xenotransplantation.

Many respondents did not favour the inoculation of nonhuman primates (NHPs) with animal source blood, cells, and/or tissues (whatever tissue source may be relevant to the proposed human clinical trials) and then monitoring the NHPs for 2 years, with or without immunosuppression. One reason was that the significant number of infectious agents harboured by NHPs would result in a high background level, clouding the interpretation. This would be particularly problematic for nonspecific tests for unknown agents. As well, there is some evidence that nonhuman primates may not be appropriate models for PERV transmission⁽²⁰⁾. The proposed standard does not address the issue of preclinical testing in animal models⁽⁸⁾, although, in general, as a prerequisite for clinical trials it is required. Given the issue of the anti-gal naturally occurring antibodies limited to humans and old world primates, studies in NHPs would be required as relevant animal models for interpretation of infectious disease risks as part of the "pre-clinical testing". In some respects, since xenotransplantation clinical trials have been formally ongoing in the U.S. since at least 1995 and may involve 50 to 100 recipients, it is not clear whether only short-term studies (days or weeks) with NHPs or follow-up over a number of years would be needed.

Most participants agreed that relevant source animal samples should be co-cultured with various relevant human indicator cell types (including PBMC and activated PBMC), tested for retroviruses and herpesviruses, examined by electron microscopy (EM), and examined for nonspecific cytopathic effects. Other means of identifying unknown pathogens (whatever is state of the art) should also be employed. Other tests include routine surveillance for infectious agents to be excluded from the herd as well as routine histopathology, hematology, and biochemistry. With regard to screening for endogenous and endemic viruses, the majority of respondents felt it was necessary to screen the herd and presumably the intended source animal(s). Many participants had trouble with the use of EM for the visual detection of virions in source animals or the herd, primarily because EM is labour intensive and relatively insensitive, in that many sections of many tissues would have to be examined in order to rule out a false negative result (sampling error). Instead, it was suggested that EM could be used in an investigation and/or on the source tissue intended for xenotransplantation, either on a biopsy specimen or an unused portion. For example, if a kidney is to be xenografted, the remaining kidney could be examined by EM.

Surrogate Markers

Several participants suggested that the following markers could be investigated as a potential surrogate for infectious disease risk: erythrocyte sedimentation rate (ESR), the PCR Enhanced Reverse Transcriptase (PERT) assay on serum, which would detect reverse transcriptase enzymatic activity characteristic of retroviruses, liver function tests, perhaps C-reactive protein, and the culture of patient materials with indicator cells (pre- and post-xenotransplantation). Some questioned whether the development of chronic fatigue or insomnia would be a useful clinical surrogate marker; others suggested that we need to develop a compendium of markers potentially related to post-xenograft complications and to establish the correlation of these to confirmed xenozoonoses. Almost no participants felt that an early cancer blood test, testing for RNAse L activity in the urine, or examination of immunosuppression of PBMC (in non-immunosuppressed recipients) would be of value as surrogate markers for xenozoonosis at this time.

Testing for Unknown Pig Infectious Agents

Some participants liked the notion of using representational difference analysis (RDA) or modified fluorescence in situ hybridization (FISH) methods to detect previously uncharacterized pig infectious agents in xenograft recipients, but others cautioned against the use of non-validated methods. Moreover, such methods are difficult to use, may be problematic, and are therefore unlikely to be suitable for routine monitoring. It may not be useful to be too specific on what tests should be employed. Some participants suggested working backwards, so that when an infectious agent is strongly suspected, then these "research methods" to identify and characterize the unknown entity can be used.

Xenozoonoses Outbreak Responses

Virtually all respondents indicated the need for SOPs at the national level for testing, sample collection, quarantine, and notification in conjunction with the federal, provincial and territorial health ministries, although the question came up as to who has the authority to invoke such measures. As well, participants recognized the need for SOPs on how to manage archived samples. Alternatively, an archive steering committee could be consulted about the use of archived materials. SOPs should also provide information on how to deal with the animal sources, including recall, lookback, and traceback procedures in the event of an outbreak. This requirement is dealt with in the Proposed Canadian Standard for Xenotransplantation.

There is a need to develop SOPs at HC for clinical holds, moratoria, inspections, and international notification in the event of an outbreak; both sets of SOPs (national and local) should be in place before clinical trials are started in Canada.

Finally, most respondents agreed that both local and national laboratories should have the facilities and resources in preparation for xenozoonotic outbreaks, including the setting up of reference assays. However, this testing may be conducted through contracts with accredited laboratories. One person suggested that, in addition, the sponsor should have SOPs for responding to a potential xenozoonotic outbreak, assuming that some clinical trials may be multicentered. Generic emergency response plans could be tailored to incorporate particulars relating to xenotransplantation infectious disease risks, although until these agents are identified and further characterized it will be difficult to provide a xenotransplantation pathogen-specific emergency response plan.

The Proposed Canadian Standard does not deal with xenozoonoses outbreaks per se, but details in generic terms what to do if an adverse reaction/suspected infection occurs⁽⁸⁾. Again, it recommends the creation of SOPs in advance on issues such as quarantine, destruction, investigation, recall, and notification, but does not provide details on how and what to investigate.

[Previous] [Table of Contents] [Next]