American Journal of Kidney Diseases
October 2002 • Volume 40 • Number 4
Controversies
in Nephrology
Protocol
biopsies should be part of the routine management of kidney transplant
recipients
|
Pro:
David Rush, MD [MEDLINE
LOOKUP] Sections |
|
|
|
|
Early protocol biopsies were instituted at our center to investigate the
pathogenesis of rapid fibrosis that we had observed in some patients. The
decision to perform these biopsies was aided by the development of
spring-loaded biopsy “guns,” which increased the safety of the procedure, and
by the standardization of allograft histopathology proposed in the Banff
conferences.
Our group has performed over 1,200 protocol biopsies. These protocol biopsies
take place in the outpatient clinic, after ultrasound localization, and
patients are discharged 3 hours after the procedure. The consent rate for
biopsy procurement is greater than 90%. There have been no graft losses;
however, we have observed two arteriovenous fistulae, and there have been three
cases of transient obstruction due to clots.
Protocol biopsies are obtained at 1, 2, 3, and 6 months posttransplant. The
early timepoints were chosen because they correspond to the period of increased
frequency of clinical rejection episodes. The 6-month biopsy was chosen to
evaluate the histological outcome and has proven to be useful as a potential
surrogate endpoint for graft survival,1 as
reported also by other investigators.2,3
Using cyclosporine-based immunosuppression, our group has found that Banff
criteria for acute tubulointerstitial rejection are present in 15% to 30% of
well-functioning kidney allografts between months 1 and 6 posttransplant, a
finding that we have referred to as “subclinical rejection.” The
tubulointerstitial inflammation observed in early subclinical rejection
represents, at least in part, a donor-specific immune response, as it
correlates with DR mismatching,4 and
with subtle recipient presensitization, as detected by flow cytometry
crossmatching.5
Most subclinical rejection is type Ia by Banff 97 criteria (ie, ai2at2), but in
one quarter of biopsies there is evidence of tubular basement rupture (type Ib;
ai2-3at3). In contrast, arteritis (av1-3) is unusual in protocol biopsy
material (Fig 1).
Fig. 1. Banff acute scores in 330
consecutive renal allograft biopsies. (A) Distribution of total Banff acute
scores. (
) Total
(330). (B) Proportion of clinical to subclinical inflammation. (
)
Clinical rejection; ()
subclinical rejection.
The importance of acute tubulointerstitial rejection as a cause of long-term
adverse graft outcomes has been downplayed in the past. However, acute tubulointerstitial
rejection that is associated with tubular basement membrane rupture has been
recently shown to lead to interstitial fibrosis6 and to
carry a similar prognosis to that of Banff type II rejection with mild to
moderate arteritis.7
Indeed, even lesser grades of tubulointerstitial rejection may lead to graft
dysfunction in kidneys procured from older donors.8
The importance of subclinical tubulointerstitial rejection in
determining long-term adverse outcomes is more controversial. However, in two
studies in which early protocol biopsies were obtained but not acted upon,
untreated subclinical rejection was correlated with an increase in interstitial
fibrosis and tubular atrophy in later protocol biopsies.9,10 In a
prospective study performed by our group,11
patients were randomized to either early (months 1, 2, and 3) or no protocol
biopsies. Both patient groups were biopsied at 6 months and were then observed
for 2 years. Patients in the biopsy arm received increased doses of prednisone
if subclinical rejection was detected. In this study, we found significantly
less interstitial fibrosis and tubular atrophy at 6 months in the biopsy group
than in the control group, despite there being no differences in the serum
creatinine. Moreover, patients in the biopsy arm of the study had a
significantly lower serum creatinine at 2 years than the controls. As might be
expected, treatment of early subclinical rejection decreased early clinical
rejections, but, perhaps more importantly, late clinical rejections, ie,
rejections occurring after 6 months, were decreased 3-fold in patients
randomized to protocol biopsies.
The finding that both histological and functional benefit can be obtained from
treating inflammation occurring in stable grafts may appear contradictory.
Indeed, the notion has been put forward, based largely on animal work, that
tubulointerstitial inflammation in such grafts may be beneficial to their
long-term acceptance. However, our group has shown that cytotoxic T cells,
their cytolytic products, and gene transcripts for proinflammatory cytokines
observed in clinically rejecting grafts are present also in subclinical
rejection, albeit in lesser amounts.12,13
Moreover, other investigators have shown that the presence of lymphocyte
markers and proinflammatory cytokine gene transcripts in grafts with stable
function correlates with subsequent functional deterioration.14
We anticipate that the reduction in clinical rejections seen with modern day
immunosuppression will be mirrored by a reduction in the prevalence of
subclinical rejection. Subclinical rejection may indeed be restricted in the
future to specific patient populations (eg, patients with high immunological
risk). Our studies to date have been in patients receiving cyclosporine-based
immunosuppressive regimens, but we are now undertaking a multicenter study of
protocol biopsies in patients treated with tacrolimus. There are no protocol
biopsy data published on kidney transplant patients receiving sirolimus or
everolimus, and these data are urgently needed. Finally, although a reduction
in the prevalence of subclinical rejection is likely to be a sensitive
indicator of the potency of a given immunosuppressive regimen, a caveat is in
order: immunosuppressive regimens that decrease the prevalence of subclinical
rejection will likely be accompanied by an increase in the rate of
opportunistic infections and perhaps malignancy.
Should protocol biopsies be a routine part of the management of renal
transplant patients? In our view, the current data balancing risk (low) and
potential benefit (high) would support a cautious “yes,” but large trials using
newer immunosuppressive regimens are required to confirm this. Moreover,
noninvasive tests sensitive enough to detect subclinical rejection need to be
developed and validated against the histology found in protocol biopsies. A
central hypothesis of our protocol biopsy work is that the episodic nature of clinical
rejection may occur on a background of deleterious subclinical inflammation
that is capable of causing renal scarring. The development of a noninvasive
test for subclinical rejection will allow for repeated surveillance of the
graft and will provide the opportunity to test this hypothesis. At present,
however, in the absence of such a test, the protocol biopsy provides
information that cannot be obtained otherwise. When performed in the early
posttransplant period, protocol biopsies can provide evidence of
immunosuppressive efficacy and at later times may identify the earliest
surrogates for long-term graft outcomes. Our view is that, because even severe
injury can be clinically inapparent in grafts with good functional reserve,
waiting for graft dysfunction to occur misses the opportunity for both
elucidation of the pathogenetic mechanisms at play and early institution of
rational treatment.
|
References |
1. Nickerson
PW, Jeffery J, Gough J, et al: The identification of clinical and
histopathological risk factors for diminished renal function 2 years
post-transplant. J Am Soc Nephrol 9:482-487, 1998
2. Seron D, Moreso F, Bover J, et al: Early protocol renal
allograft biopsies and graft outcome. Kidney Int 51:310-316, 1997
3. Nicholson
ML, Bailey E, Williams S, Harris KP, Furness PN: Computerized histomorphometric
assessment of protocol renal transplant biopsy specimens for surrogate markers
of chronic rejection. Transplantation 68:236-241, 1999
4. Rush
D, Jeffery J, Gough J, et al: Predicting rejection: Is early diagnosis
achievable and important? Graft 2:S31-S35, 1999
5. Karpinski M, Rush D, Jeffery J, et al: Flow cytometric
cross-matching in primary renal transplant recipients with a negative
anti-human globulin enhanced cytotoxicity cross-match. J Am Soc Nephrol
12:2807-2814, 2001
6. Bonsib
SM, Abul-Ezz SR, Ahmad I, et al: Acute rejection-associated tubular basement
membrane defects and chronic allograft nephropathy. Kidney Int 58:2206-2214,
2000
7. Minervini
MI, Torbenson M, Scantlebury V, et al: Acute renal allograft rejection with
severe tubulitis (Banff 1997 grade IB). Am J Surg Pathol 24:553-558, 2000
8. De Fijter JW, Mallat MJK, Doxiadis IIN, et al: Increased
immunogenicity and cause of graft loss of old donor kidneys. J Am Soc Nephrol
12:1538-1546, 2001
9. Legendre C, Thervet E, Skhiri H, et al: Histologic
features of chronic allograft nephropathy revealed by protocol biopsies in
kidney transplant recipients. Transplantation 65:1506-1509, 1998
10. Nankivell BJ, Fenton-Lee CA, Kuypers DR, et al: Effect of
histological damage on long-term kidney transplant outcome. Transplantation
71:515-523, 2001
11. Rush D, Nickerson P, Gough J, et al: Beneficial effects
of treatment of early subclinical rejection: A randomized study. J Am Soc
Nephrol 9:2129-2134, 1998
12. Lipman M, Shen Y, Jeffery J, et al: Immune-activation
gene expression in clinically stable renal allograft biopsies: Molecular
evidence for subclinical rejection. Transplantation 66:1673-1681, 1998
13. Grimm PC, McKenna R, Nickerson P, et al: Clinical
rejection is distinguished from subclinical rejection by increased infiltration
by a population of activated macrophages. J Am Soc Nephrol 10:1582-1589, 1999
14. Kirk AD, Jacobson LM, Heisey DM, Radke NF, Pirsch JD,
Sollinger HW: Clinically stable human renal allografts contain histological and
RNA-based findings that correlate with deteriorating graft function.
Transplantation 68:1578-1582, 1999
Protocol
biopsies should be part of the routine management of kidney transplant
recipients
|
Con: Daniel R. Salomon, MD Sections |
|
|
|
|
The thesis is that a protocol-driven kidney transplant biopsy in a patient with
essentially stable renal graft function will yield a result directing
immunosuppressive therapy and enhance long-term transplant survival and
function. While I strongly support continued research in this area, I do not
think that this premise has been proven or sufficiently refined to be ready for
general implementation. A biopsy-based treatment strategy is both invasive and
expensive but, even more critically, requires a significant intensification of
baseline immunosuppression for those patients triggering the treatment
criteria. Thus, I will argue that the burden of proof still remains on us.
Let me begin by stipulating that the immunological underpinnings of the thesis
are soundly based on work in both experimental animal models and clinical
practice. Renal transplant function is an inadequate measure of ongoing tissue
injury due to rejection. The kidney's ability to hypertrophy and compensate for
loss of functioning nephrons is one mechanistic explanation for the disparity
between measured function and ongoing tissue damage. Certainly in many patients
with chronic rejection, by the time we become aware of renal dysfunction the
biopsy presents extensive interstitial and vascular injury and fibrosis.
Moreover, a series of publications from Rush and colleagues1-4 have
provided good evidence that protocol biopsies can indeed reveal histological
changes consistent with acute and/or chronic rejection in patients with stable
and, by clinical standards, ostensibly “normal,” renal transplant function. My
only comment is that over a decade ago, in the context of recognizing this
disconnect between renal function and underlying rejection, a number of groups
advocated more precise measurements of renal transplant function. Strategies
included true clearances, nuclear medicine scans, and even protein-loading to
test whether loss of the expected and compensatory increase in clearance would
be a more sensitive marker of function. Frankly, the complexity and expense of
these clearance studies never fit well with busy transplant clinics, and the
practice was abandoned by most centers. If the option now is a renal biopsy and
preemptive treatment, perhaps more sensitive tests of renal function should be
reconsidered again as a surrogate marker of subclinical rejection.
The fact is that all immunosuppression is imperfect and, thus, essentially all
transplants generate an alloimmune response. So how do we interpret
histological criteria for rejection in a protocol biopsy of a transplant with
stable function, when we admit that all transplants will generate an immune
response? In other words, theoretically, a biopsy at some early point
posttransplant must show this immune response. Indeed, most of our current
paradigms for the induction and maintenance of tolerance require an immune
response. Thus, we also have to consider the possibility that short-term
treatment of acute immune responses might alter long-term allograft survival or
limit our opportunities to reduce immunosuppression later.
Clearly, the simple presence of T cells, B cells, and/or macrophages is not
sufficient to conclude rejection is present and requires treatment. Can we make
a distinction based on time posttransplant? For example, an immune response at
3 weeks might be “normal,” but at 12 weeks it might be indicative of
insufficient immunosuppression? Perhaps adding tissue injury criteria to the
equation would help, although a number of renal injury mechanisms are operating
in most transplant patients, including drug toxicities, underlying diseases
such as diabetes, and acquired pathologies such as hypertension and
compensatory hyperfiltration. Of course, all inflammatory cell infiltrates are
not equivalent. Thus, many groups have suggested the use of various markers to
identify T-cell subsets, cell activation markers, and more recently polymerase
chain reaction-based identification of proinflammatory transcripts in tissue or
urine5
or novel technologies such as urine spectroscopy.6 While
all of these strategies have considerable promise, my point is that all of them
remain largely experimental and that the number of open questions is
significant. Despite the absolutely sound reasoning behind using the latest
molecular technologies to refine our interpretations of the biopsy and/or
monitor the immune response in transplant patients, the fact is that predicting
response to therapy, adequacy of immunosuppression or long-term graft survival
remains difficult. Thus, I believe we need more evidence before adopting a
clinical strategy where the result of a protocol biopsy with standard
histological criteria for rejection is the basis for predicting long-term graft
function or initiating acute rejection therapy in patients with stable kidney
transplant function.
A critical distinction made by Rush and others is between histological changes
of acute and chronic rejection in a protocol biopsy.6 I
acknowledge that finding chronic rejection-like changes in a protocol biopsy at
6 months (ie, increased interstitial fibrosis score) would be of concern and
make it easier to rationalize a potentially effective preemptive treatment.
However, this raises a different set of questions. First, what is the effective
treatment for chronic rejection? Given that there is no good answer, the
presumption has to be that “chronic rejection,” as recognized in the protocol
biopsy, is actually the consequence of ongoing acute rejection. Thus, it is the
acute rejection component that must respond to treatment. In my opinion, the
existing data do not prove this hypothesis, although the hypothesis is sound
and the clinical trial designs to accomplish this proof are reasonably
straightforward.
Secondly, I remain uncertain about exactly what proponents of protocol biopsies
are planning for general implementation. Particularly, if the objective is to
detect this variety of subclinical chronic rejection, is the process of
protocol biopsies intended to extend to the lifetime of the transplant? If so,
what is the optimal biopsy timing and frequency in patients over 1 year
posttransplant? Is it going to be useful to track changes in serial protocol
biopsies before making an intervention or will decisions be based on a single
biopsy result? How can an iterative analysis of serial biopsies be done in a
standardized fashion? In real practice, is a yearly protocol biopsy practical?
More importantly, will biopsies done years posttransplant consistently provide
enough warning of underlying acute rejection that treatment will prevent, or at
least delay, impending chronic rejection? Alternatively, the argument could be
that the future of a renal allograft is “written” in the first year, which
confines the universe of time for protocol biopsies. Again, I find this to be
an interesting hypothesis, but one that will require prospective multicenter
trials to establish.
Finally, much evidence suggests that chronic rejection is a complex mix of
renal injury mechanisms. If we set aside the underlying acute rejection
hypothesis for a moment, it must be proven that chronic rejection detected in
protocol biopsies provides an indication of a therapy-responsive injury. That
will also require establishing the window of time posttransplant within which
such a conclusion is correct. In other words, chronic rejection histology
identified 5 years posttransplant is rarely therapy-responsive. Moreover, many
would argue that the underlying mechanisms are probably very different than
what is happening in the first year, regardless of the similar histology. Thus,
is there something mechanistically different about chronic rejection defined
histologically in the first year as compared to that discovered relatively
late, or is the key difference the early detection of an underlying and
treatable acute rejection? My point is that making a distinction between acute
and chronic rejection is very appropriate, but the arguments for a
therapy-responsive acute rejection detected by a protocol biopsy in the first
year posttransplant are much easier to understand than are those for chronic
rejection, unless the latter is simply a surrogate marker for a subclinical
acute rejection.
If a protocol biopsy in a patient with stable renal function detects sufficient
evidence of rejection, then the next step is the intention to treat. Every
patient is on an immunosuppressive protocol. Therefore, the evidence of
rejection, particularly with tissue injury, means that this immunosuppression
is not adequate for that patient. So why is it correct to give a short course
of high-dose corticosteroids without any change in the underlying
immunosuppression? It would seem to be inescapable logic that such an
intervention might successfully ameliorate an ongoing, subclinical acute
rejection and leave the transplant open to a relentless, low-grade, chronic
immune attack that will shorten the life of the graft and eventually be
recognized as chronic rejection. I realize that many would claim that a bolus
of intense acute rejection therapy makes some permanent change in the immune
response. I would point out two decades of similar arguments for induction
antibody therapies that they should have produced tolerance. I would also note
that acute rejection episodes, particularly those 
6 months
posttransplant, are a highly correlated risk factor for reduced long-term graft
function and survival despite their good immediate response to antirejection
therapy.
In fact, it is interesting that newer immunosuppressive drug regimens have
significantly reduced the incidence of acute rejection but had much less impact
on long-term graft survival. This point was actually raised in one of the
studies by Nickerson et al,4 in
which they suggest that despite the reduction in clinical acute rejection with
newer drug regimens, there is not a parallel decrease in the subclinical acute
rejection detectable by protocol biopsy. Of course, the standard practice
remains to steadily reduce immunosuppression as a function of time
posttransplant, such that levels of immunosuppression typically used at 1 year
would never be considered adequate to prevent acute rejection at 1 month. This
is often even more the case with the newer agents, in which the thinking has
been that because these agents are more “powerful,” we can now discontinue
steroids or avoid long-term immunosuppressive complications by reducing or
stopping key drugs after 6 months without rejection. I have never been
comfortable with this logic in the context of my belief that maintaining
adequate long-term immunosuppression is important and given the fact that we
still do not have a “test” to measure the adequacy of immunosuppressive
therapy. Thus, my point is that more work is required to establish the therapy
options for protocol biopsy results. I am not ready to commit to a strategy
based primarily on a short-term intensification of baseline immunosuppression
with corticosteroids or even anti-T-cell antibodies. Why not simply increase
the baseline immunosuppression after a positive protocol biopsy and repeat the
biopsy in 2 months? Maybe it will be necessary to treat and increase baseline
immunosuppression to get the best result.
Another issue is that multicenter, prospective trials have not been done yet
with protocol biopsies. Despite the significant value of the Banff criteria in
standardizing pathological criteria for rejection, a proof for the utility of
protocol biopsies must consider the complex differences in clinical practice,
patient selection, induction therapies, baseline immunosuppression, long-term
immunosuppression, and experience that marks the reality of clinical programs.
A number of perfectly reasonable clinical paradigms with good proof of
principle studies have not been proven in multicenter trials. In some cases
this is because the incremental value of the therapy is not great enough to
rise above the greater noise inherent in a large and multicenter trial. In
other cases, the premise supported in early, limited trials just turns out to
be wrong. Moreover, the outcome parameters tested in a multicenter trial of
protocol biopsies must include a minimum of 5-year graft function and survival.
In conclusion, it is probable that protocol biopsies in a subset of kidney
transplant patients with normal and stable renal function will reveal evidence
of a therapy-responsive and subclinical acute rejection. The challenges are to
refine the criteria and timing for the biopsies, to determine the best
treatment strategy, and to demonstrate the principle in practice by a
well-designed, prospective, multicenter trial. The evidence must prove that
preemptive therapy based on a protocol biopsy actually improves long-term graft
survival and function significantly. If the benefit is demonstrable but only
marginal, then the wisdom of subjecting all patients to multiple biopsies and
changes in immunosuppression will need to be considered carefully. The
possibilities of integrating the latest technologies for molecular diagnosis of
rejection must also be considered and compared to the biopsy, alone, for
utility and predictive value. In other words, even if a protocol biopsy can
predict long-term graft function or survival, if noninvasive alternatives can
be developed with equal utility, then they will be preferred. Nonetheless, the
clinical validation of such molecular approaches will owe much to the parallel
biopsy material. If so, the protocol biopsy will take its place in history as
the procedure that allowed validation of these preferred strategies.
|
References |
1. Rush
D, Nickerson P, Gough J, et al: Beneficial effects of treatment of early
subclinical rejection: a randomized study. J Am Soc Nephrol 9:2129-2134, 1998
2. Lipman ML, Shen Y, Jeffrey JR, et al: Immune-activation
gene expression in clinically stable renal allograft biopsies: molecular
evidence for subclinical rejection. Transplantation 66:1673-1681, 1998
3. Nickerson
P, Jeffrey J, Gough J, et al: Identification of clinical and histopathologic
risk factors for diminished renal function 2 years posttransplant. J Am Soc
Nephrol 9:482-487, 1998
|
4. Nickerson P, Jeffrey
J, Gough J, et al: Effect of increasing baseline immunosuppression on the
prevalence of clinical and subclincial rejection: A pilot study. J Am Soc
Nephrol 10:1801-1805, 1999 |
|
|
6. Rush D, Somorjai R, Deslauriers R, et al: Subclinical
rejection-a potential surrogate marker for chronic rejection-may be diagnosed
by protocol biopsy or urine spectroscopy. Ann Transplant 5:44-49, 2000