icon star paper   Hepatitis C Articles (HCV)  
Back grey_arrow_rt.gif
 
 
Hepatitis C, interferon, and risk of rejection after liver transplantation
 
 
  EDITORIAL
Liver Transplantation
Volume 10, Issue 7, July 2004
 
Didier Samuel *Centre H╚patobiliaire, H┘pital Paul Brousse, Villejuif, France
 
Hepatitis C virus (HCV)-related end stage cirrhosis is currently the leading indication for liver transplantation in Europe and in the United States. HCV recurrence after transplantation is almost universal and 60 to 90% of patients will develop lesions of chronic hepatitis C on the graft. Recent data confirm that HCV infection impairs patient and allograft survival. The course of HCV graft disease is accelerated in transplant recipients compared with immune-competent patients, with reported 5-year cirrhosis rates around 10 to 20% and up to 28 to 40% in some series. This yields excess a risk of death or retransplantation for liver failure at 10-15 years posttransplant.
 
It appears now that the survival at 5-10 years is impaired in HCV-positive recipients in comparison to non-HCV recipients, due in part to the development of liver cirrhosis and of liver failure after liver transplantation. Given this risk, it appears important to offer antiviral therapy to liver transplant patients who develop a recurrence of chronic hepatitis C.
 
Several approaches are under evaluation: the first option is to start the treatment just after transplantation as a preemptive therapy, the advantage is that the viral load is low at that time; however, tolerance of patients is poor in the immediate posttransplant period, and risk of rejection and of sepsis are high. The second option is to treat at time of acute hepatitis (in general during posttransplant months). The third option is to start the treatment at time of chronic hepatitis on the graft; in this latter situation the groups are heterogeneous, since some patients are treated during the first posttransplant year, some much later, with major differences both in the immunosuppressive drugs given to these patients, and in the severity of graft hepatitis.
 
The type of antiviral treatment used is now mostly the combination therapy interferon plus ribavirin. Indeed, results with interferon alone have been disappointing in term of virologic response, several studies using nonpegylated interferon alone have reported a sustained virologic response rate below 10%.
 
The combination interferon alpha-2b plus ribavirin combination therapy gave promising results in liver transplant patients with chronic hepatitis C on the graft, achieving sustained virologic response rate of 20 to 35%. It is not well known if the use of pegylated interferon in the combination therapy will further improve these results. However, the first reported results of posttransplant treatment with pegylated interferon plus ribavirin gave promising results, achieving sustained virologic response in 25 to 40% of the patients.
 
The antiviral treatment is particularly important in the management of these patients, since progression of fibrosis can probably be stopped after sustained virologic response. However, there are some controversies on the long-term outcome of the graft histology after sustained virologic response, and some authors have described a decreased activity grade and an absence of true improvement in the fibrosis score at least during the first years posttreatment.
 
Patients with HCV reinfection of the graft after liver transplantation differ from immune-competent patients in several ways: their viral load is very high, and most transplant patients in Europe are infected with genotype 1 - both factors predictive of a lower virologic response rate. Despite the absence of certitude on the possibility of fibrosis regression, patients with sustained virologic response have normal liver enzymes level, a dramatic decrease of the activity score, and probably a much better long-term outcome in comparison to the patients with ongoing infection.
 
For these reasons, the antiviral treatment is now fully part of the overall therapeutic strategy posttransplantation. The timing, the duration of treatment, the use of pegylated interferon instead of nonpegylated interferon, and the optimal dosage of ribavirin are still matter of debate. It has been clearly shown that in the absence of treatment there is no decrease of viral replication, no spontaneous clearance of the virus, and there is a progression of liver graft disease.
 
The drawbacks of the treatment are a rather disappointing virologic efficacy and a poor tolerance. Indeed, the treatment with interferon plus ribavirin is poorly tolerated and tolerance appears lower than in immune-competent nontransplant patients.
 
One of the main complications reported in comparison to nontreated patients is anemia; this is due to an increase toxicity of ribavirin in these patients with frequently impaired renal function, and with an impaired regenerative bone marrow. For this reason, it was necessary to introduce ribavirin at lower dosage than in nontransplant patients, to add erythropoietin administration in some patients, and to decrease the dosage of ribavirin, reducing the efficacy of the antiviral treatment. The other complications are similar to nontransplant patients: leukopenia, thrombocytopenia, depression, and irritability.
 
The risk of rejection in transplant patients treated with interferon is specific to transplant recipients, but remains a matter of debate. It is well known that interferon can provoke acute or chronic rejection of the renal allograft in renal transplant patients; there are numerous reports of irreversible renal failure due to rejection in renal transplant recipients treated with interferon.
 
Until recently, interferon was still considered as a contraindication in renal transplant patients. The risk and severity of rejection due to interferon in liver transplant recipients are still controversial. Despite some reports on the possible occurrence of acute or chronic rejection in liver transplant recipients, this complication is not always recognized, and the role of interferon is controversial. This may be due to several factors: the risk of rejection due to interferon is probably lower in liver transplant recipients than in renal transplant recipients, the liver being considered as more resistant to rejection than the kidney. Not all patients have been biopsied in the presence of abnormal liver enzymes during interferon, and the risk of rejection is probably underestimated. The risk of rejection with interferon might be dependent of the posttransplant interval and of the level of immunosuppressive drugs.
 
It is unclear whether the risk of rejection due to interferon is modified by the use of pegylated interferon or by the adjunction of ribavirin. Pegylated interferon may theoretically increase the risk of rejection, due to its greater efficacy, its different pharmacokinetics, and its renal clearance, which is frequently decreased in liver transplant patients. On the contrary, it has been suggested that the adjunction of ribavirin might decrease the risk of rejection, since ribavirin has immune modulatory action. This has been suggested by the lower number of reported cases of rejection in liver transplant patients treated with interferon plus ribavirin therapy than with interferon alone and by a recent report of 4 cases of renal transplant patients treated successfully for acute hepatitis C without occurrence of rejection.
 
The report by Saab et al. in this issue of Liver Transplantation highlights the risk of rejection in liver transplant patients treated with combination pegylated or nonpegylated interferon plus ribavirin. Five out of 41 patients experienced acute or chronic rejection episodes during combination treatment, leading in several cases to discontinuation of the antiviral treatment. Several points appeared from this study: 1) the cases of rejection are well documented; 2) the risk of rejection appeared as relatively low (4/41, 10%) but remains high if the late period after transplantation is taken into account; 3) the consequences of these episodes of rejection can be severe, since some patients died or were retransplanted for chronic rejection. Most of these patients experienced rejection at long-term posttransplantation, at a time where the risk of rejection is usually low reinforcing the deleterious role of interferon in these cases; 4) rejection occurs relatively rapidly after the start of the treatment, during the first 3 months in 4 out of 5 cases.
 
The immunosuppressive regimen was relatively low in most patients at the start of the treatment. In the report of Feray et al. with interferon alone, the rate of rejection was higher (35%), mostly represented by severe chronic rejection, and was significantly higher than pair-matched untreated HCV liver transplant patients. In the report of Saab et al., patients experienced both features of acute and chronic rejection. The mechanisms of action of interferon on rejection are not well known, however, it seems to activate human leukocyte antigen class II expression on biliary ducts, triggering lymphocyte infiltrate or disappearance of interlobular biliary ducts. The fact that ribavirin may decrease the risk of rejection of interferon has been suggested but has not been really proven. In a randomized study with interferon plus ribavirin, we observed 1/28 episodes of rejection in the treated group vs. 0/24 in the nontreated group.
 
Why these differences in the prevalence and the severity of rejection episodes during interferon treatment among different studies? First, the delay when initiating the treatment after transplantation is important, the risk of rejection being higher in the first months posttransplant. Second, the immunosuppression dosage used at the time antiviral treatment begins may interfere positively or negatively with the risk of rejection. Third, the prevalence of rejection can be underestimated particularly in nonresponding patients, the increase in liver enzymes being considered falsely as due to hepatitis C in the absence of liver biopsy. Fourth, the use of pegylated vs. nonpegylated interferon and the combination with ribavirin might modify the prevalence of rejection. Indeed, pegylated interferon might increase the risk of rejection due to its long-lasting effect and its renal clearance in patients with impaired renal function. Recently, a prevalence of 25% of mild rejection episodes was reported in a series of 20 patients treated with pegylated-interferon plus ribavirin.
 
In conclusion, from these reports, we can conclude that while there is a risk of rejection during interferon-based treatment for hepatitis C recurrence after liver transplantation, the prevalence and the severity of rejection remains matters for debate. The risk of rejection might be higher in patients receiving pegylated interferon, however this has not yet be proven. The decreased risk of rejection due to the combined use of ribavirin has been suggested, but not proven. Finally, these reports point out the need to perform a liver biopsy before starting the antiviral treatment, to eliminate any ongoing histological rejection, to maintain a sufficient level of immunosuppression (which should be balanced with the need to decrease viral replication), to closely monitor liver tests during antiviral treatment, and to perform a liver biopsy in case of increased liver enzymes. The risk of rejection during interferon-based treatment should not be ruled out and should be evaluated in further studies.
 
Outcomes of acute rejection after interferon therapy in liver transplant recipients
 
Sammy Saab 1 2 *, Denise Kalmaz 1, Nupoor A. Gajjar 3, Jonathan Hiatt 2, Francisco Durazo 1 2, Steven Han 1 2, Douglas G. Farmer 2, R. Mark Ghobrial 2, Hasan Yersiz 2, Leonard I. Goldstein 2, Charles R. Lassman 3, Ronald W. Busuttil 21Department of Medicine, Dumont-UCLA Liver Transplant Center, Los Angeles, CA2Department of Surgery, University of California at Los Angeles, Los Angeles, CA3Department of Pathology, University of California at Los Angeles, Los Angeles, CA
 
ABSTRACT
 
Interferon alfa has been increasingly used against recurrent hepatitis C (HCV) disease in post-liver transplant (LT) recipients. A serious potential adverse effect is acute rejection. We reviewed our experience using interferon-based therapy (interferon or pegylated interferon with or without ribavirin) for treating recurrent HCV in LT recipients.
 
Forty-four LT recipients were treated with interferon for recurrent HCV. Five of the 44 patients developed acute rejection during interferon-based therapy. These 5 patients started treatment of 42.4 ▒ 33.89 months (mean ▒ SD) after LT.
 
Mean (▒ SD) histological activity index and fibrosis scores before initiating antiviral therapy were 8.8 (▒ 1.92) and 2.6 (▒ 0.55), respectively. Patients were treated for 3.3 ▒ 2.28 months (mean ▒ SD) prior to rejection. At the time of rejection, HCV load was not detectable in 4 of the 5 recipients. All 5 patients had tolerated interferon therapy, and none had stopped therapy because of adverse effects. The rejection was successfully treated in 3 patients. In 2 of those 3 patients, cirrhosis eventually developed. In the 2 patients who did not respond to rejection treatment, immediate graft failure occurred, leading to re-LT in 1 patient and death from sepsis in the other.
 
In conclusion, the results indicate that further studies are needed to assess the safety of interferon in LT recipients. Interferon-based therapy may lead to acute rejection and subsequent graft loss and should therefore be used with caution. Treated recipients may also develop progressive cirrhosis despite achieving a sustained virological response.
 
BACKGROUND
 
Hepatitis C (HCV) is the leading indication for liver transplantation in the United States. The number and proportion of individuals receiving transplants for HCV will likely grow as the cohort infected several decades ago start seeking medical care. Recurrent HCV viremia is universal, with approximately 90% of individuals undergoing transplantation for HCV having detectable virus soon after transplantation.Histological disease is seen in 50% of recipients within 2 years. After 5 years, up to 30% of patients transplanted for HCV develop cirrhosis.The rate of decompensation after developing cirrhosis is approximately 42% after the first year in the posttransplant setting. Although earlier studies demonstrated no impact of recurrent HCV on patient survival, more studies suggest HCV may indeed negatively impact patient and graft survival.
 
As in nontransplant patients and the general population, the treatment of recurrent HCV disease after liver transplantation involves the use of interferon, which is believed to have both antiviral and immunomodulatory effects against HCV. However, in the posttransplant setting, antiviral therapy has been limited by relative ineffectiveness, poor tolerability, and adverse effects. Long-term response to combination interferon and ribavirin in transplant recipients is estimated to be 20% in patients treated for recurrent HCV. Efficacy is increased when pegylated interferon is used in combination with ribavirin, but it is still believed to be around 30%. Tolerance to interferon therapy after transplantation also appears to be less than in the general population, with exaggerated adverse effects including anemia and leukopenia.
 
A serious complication of interferon in organ transplant recipients is the precipitation of acute rejection and subsequent graft loss. Overall, the majority of acute rejection episodes after liver transplantation occur within 90 days of surgery and are seen in up to 75% of recipients, regardless of immunosuppressant regimen. However, acute rejection after liver transplantation in the setting of interferon therapy is not a well-documented phenomenon. This complication has been seen in renal transplant recipients and is believed to be due to the immunomodulatory effects of interferon. As a result, interferon is avoided in renal transplant recipients except in the setting of severe liver complications secondary to HCV. The incidence of serious acute rejection in renal transplant recipients treated with interferon ranges between 15 and 64%. Although interferon is used in liver transplant (LT) recipients to treat recurrent HCV, few data exist on its association with acute rejection. Feray et al. first reported chronic rejection in LT recipients treated with interferon for recurrent HCV.
 
In the current study, we sought to examine the incidence and clinical outcome of patients who developed acute rejection while on interferon therapy.
 
A chart review was performed on patients who underwent liver transplantation and were treated with interferon-based therapy for recurrent hepatitis C at the University of California Los Angeles Medical Center. We searched an administrative database and identified 44 recipients who received interferon, interferon and ribavirin, pegylated interferon, and pegylated interferon and ribavirin between January 1998 and May 2003. We collected data on recipient variables: sex, comorbidities, age at transplant, year of transplant, immunosuppressants, duration of treatment, and hematological, biochemical and virological results. We also collected data on treatment initiation date, date of cessation of interferon therapy, and reason for cessation (if course not completed), along with any interferon dose modifications made during treatment. In addition, the immunosuppression regimen, including corticosteroids, cyclosporine, tacrolimus, azathioprine, and mycophenolate mofetil, was followed along with any changes made over the course of interferon therapy. Side effects including anemia, leukopenia, and thrombocytopenia were recorded, along with any interferon dose modifications and/or use of growth factors such as erythropoietin and filgrastim. Psychological adverse effects such as depression and insomnia were also documented.
 
The immunosuppression regimens after LT consisted of combinations of cyclosporine, prednisone, tacrolimus, azathioprine, and mycophenolate mofetil. According to our protocol, patients initially received 1 gram of intravenous methylprednisolone the day of transplantation; this was then tapered to 20 mg/day over 1 week. Oral prednisone was subsequently started at 20 mg/day and tapered as tolerated. Beginning in 1994, our institution established a regimen consisting of tacrolimus for maintenance immunosuppression. Generally, our protocol for posttransplantation target immunosuppression levels for tacrolimusconsists of 10 to 12 ng/mL at 1 to 3 months; 8 to 10 ng/mL at 3 to 6 months; 5 to 8 ng/mL at 6 to 12 months; and 2 to 5 ng/mL at greater than 12 months. For cyclosporine, the target values are 250 to 300 ng/mL at 1 to 3 months; 175 to 250 ng/mL at 3 to 6 months; 100 to 175 ng/mL at 6 to 12 months; and 50 to 100 ng/mL at greater than 12 months. Immunosuppression levels were checked bimonthly to monthly, with changes in doses made to reach target ranges. We collected immunosuppression values before and during antiviral therapy, as well as at the time of rejection.
 
The indication for performing liver biopsies in all transplant recipients was elevated liver-associated tests found on routine lab work. Protocol biopsies are not performed at our institution. Liver biopsies were read with an experienced LT pathologist. The histological activity index and fibrosis score, and the diagnosis of rejection, were determined according to established criteria. Criteria for beginning interferon therapy included detectable HCV RNA and a fibrosis score of 2 or more, and/or elevation of liver-associated tests at least twice the upper limit of normal.
 
Histological criteria used for the diagnosis of rejection included (1) portal inflammation composed of a mixed infiltrate (lymphocytes, plasma cells, eosinophils, and neutrophils); (2) bile duct inflammation and damage (lymphocytes within bile duct epithelium); and (3) venous endotheliitis demonstrated by lymphocytes infiltrating the endothelial and lifting up luminal endothelial cells. Rejection severity was based on the number of portal tracts affected by the infiltrate, the degree of damage, and the number of bile ducts damaged. There have been no changes in the histological diagnostic criteria for rejection over the study period.
 
Author Discussion
 
Interferon therapy is a widely accepted treatment for LT recipients with recurrent hepatitis C. However, our study shows that interferon therapy in this population is not without serious risks. Acute rejection occurred in 5 of 44 patients (11.4%) treated with interferon at our institution. Two of the 5 patients developed graft loss as a result of steroid-resistant rejection, with 1 requiring a second LT and the other expiring from sepsis. The remaining 3 patients responded to steroid pulsation and/or additional immunosuppressant agents and recovered completely from the rejection episode. However, 2 of these 3 patients subsequently developed cirrhosis. This rapid progression of their recurrent HCV may be related to accelerated viral replication after intensified immunosuppressive therapy.
 
In the current era of liver transplantation, graft loss due to acute rejection is rare. Most episodes of acute rejection occur within 3 months of surgery and are responsive to steroids. In a cohort of 55 LT recipients treated for early acute cellular rejection, there were no deaths or retransplantation. Late acute rejection occurs after more than 180 days posttransplantation and has been reported at rates of 7 to 10% of all transplant patients. Late acute cellular rejection tends to be refractory to medical treatment, particularly if the indication for transplantation was viral hepatitis. Nevertheless, graft loss or death from late acute rejection in a series of 122 patients was still uncommon, occurring in 14 recipients. In the present study, acute rejection developed up to 83 months after liver transplantation, and 2 of the 5 patients who developed late acute rejection eventually experienced graft loss leading to retransplantation in 1 and death in the other.
 
Interferon has not previously been associated with an increased risk of acute rejection in post-LT patients. Several studies have supported the long-term efficacy and safety of interferon in this population. However, many of the published studies included small study groups. Jain et al. published a study of the safety and efficacy of interferon therapy. The study included LT recipients with either hepatitis B or C. They concluded that the risk of acute rejection was not increased by the introduction of interferon alfa. However, the patients who received interferon were given higher levels of prednisone during the trial than the controls. Feray et al. showed that chronic rejection can occur in up to 35% of patients treated with interferon.
 
Although the relationship between interferon therapy and acute rejection in LT recipients has not been well documented, this association has been well established in the renal transplant population. Interferon was initially studied in these patients as prophylaxis against cytomegalovirus (CMV) infection after transplantation. In a study by Kovarik et al., there was a higher rate of rejection and a significantly higher number of HLA-DR mismatches in the group of patients who received interferon compared with the placebo group. The study was discontinued because of ethical concerns over the high rate of rejection after treatment with interferon in this population. Subsequent studies of renal transplant patients on interferon have also shown increased rates of steroid-resistant acute vascular rejection episodes. As a result of this data, a recent review recommends that interferon therapy not be used in renal transplant recipients except those with fibrosing cholestatic hepatitis. It is unclear why renal transplant grafts appear to be more vulnerable to rejection episodes after interferon than LT grafts. HLA compatibility has been shown to influence the outcome of renal transplants to a greater degree than liver transplants.
 
As a result of the immunomodulatory and immunostimulatory effects of interferon, several mechanisms behind the induction of acute rejection have been postulated. One hypothesis is that interferon alfa may enhance the expression of HLA-DR antigens on donor hepatocytes, thus initiating a vigorous immune response and subsequent episode of rejection. This has been supported by in vivo studies showing that HLA-DR expression is enhanced in response to human interferon. Alterations in class II histocompatibility expression in transplanted tissue by interferon alfa have also been seen in a model of heart transplantation in rats. In an analysis of human liver grafts after transplantation, when rejection was the reason for graft failure, bile duct epithelial cells were found to be HLA-DR positive, as were hepatocytes in severe rejection episodes. The bile duct epithelia of donor specificity is thought to become the dominant class II antigenic stimulant and contribute to the vanishing bile duct syndrome seen in rejection. Dousset et al. reported the observation of acute vanishing bile duct syndrome in 2 of 5 patients treated with interferon: 1 of those 2 patients required retransplantation;the other expired. This hypothesis of enhancement in the expression of HLA-DR antigens by interferon can also be supported by recent evidence that pegylated interferon may be more likely to cause rejection than traditional interferon alfa. In the present study, 4 of the 5 patients who experienced rejection received pegylated interferon. A possible mechanism for this increased rejection risk is related to the extended half-life and increased serum concentration of pegylated interferon. These properties could make it more likely to increase HLA expression, and therefore rejection, than interferon.
 
Another proposed mechanism for the increased risk of interferon-associated acute rejection may be decreased levels of immunosuppressants as a result of HCV clearance during antiviral therapy. It is suggested that antiviral therapy improves hepatocyte microsomal function, which leads to decreased immunosuppression levels. However, in our study, only 1 patient who developed rejection had a decreased mean immunosuppression level while on interferon. In addition, 3 patients who did not develop acute rejection had immunosuppression levels below our recommended protocol while on antiviral therapy. Decreased immunosuppression levels may be important; however, other factors are likely required for the development of acute rejection.
 
Although the exact mechanism is unknown, interferon is believed to lead to HCV clearance by inhibiting viral replication and enhancing the immune response against the hepatitis virus. The use of an immunomodulator such as interferon with immunosuppressants such as cyclosporine, tacrolimus, and prednisone raises concerns about possible interactions. These medications appear to inhibit T-cell activation and may inhibit interferon production. This has been suggested by the observation of low levels of circulating interferon alfa in renal transplant recipients on immunosuppressant therapy. These studies raise concerns about the use of interferon in a transplant and immunosuppressed population. The effects of the additional use of growth factors such as filgrastim to improve tolerance of interferon therapy and avoid leukopenia also have yet to be studied. In addition, the possible protective effect of ribavirin against acute rejection as an adjunct to interferon remains to be explored. In the present study, 2 of three patients treated with pegylated interferon monotherapy experienced acute rejection. Early studies have suggested that ribavirin may have immunosuppressive effects. Ribavirin may therefore provide additional immunosuppressive qualities that provide protection from rejection in addition to its role in antiviral therapy.
 
Another important consequence of acute rejection after treatment with interferon observed in our study, aside from graft loss from resistant rejection, is the acceleration of the course of recurrent hepatitis C. Two patients who responded to increased immunosuppression and recovered from rejection subsequently experienced progressive cirrhosis and graft failure. With intensive immunosuppression, liver disease progression may be accelerated. Therefore, the impact of treatment with interferon on graft survival after achieving serum virological response remains to be studied.
 
Also observed in our study were the features of mild acute rejection in the biopsies of 3 of the study patients prior to the initiation of interferon therapy. Two of these patients subsequently experienced graft failure after developing acute rejection while on interferon therapy. Distinguishing between graft hepatitis and acute cellular rejection can be difficult because the histological features of the 2 conditions can overlap. For example, portal infiltrate can be found in both rejection and recurrent HCV. However, in the setting of elevated quantitative HCV RNA levels and significant biopsy-proven elements of recurrent HCV after more than 7 months posttransplantation, the decision was made to treat with antiviral therapy. Gottschlich et al. suggest that quantitative HCV RNA levels can be useful in distinguishing between recurrent HCV and rejection. After treatment with interferon, the biopsy results of the 2 patients became diagnostic of acute rejection and interferon was discontinued. Therefore, it is difficult to determine whether or not interferon accelerated the process of acute rejection in these patients. However, while some histological features of recurrent hepatitis and rejection may overlap, it should be noted that definitive evidence for acute rejection, such as endotheliitis, was not seen prior to interferon therapy in any of the patients.
 
Limitations of our study include the lack of protocol biopsies accounting for possible unrecognized mild rejection and the retrospective nature of the analysis. An additional limitation is that the rejection associated with clearance of HCV infection may have been unrelated to interferon therapy. Indeed, Doughty et al. reported on a single patient who cleared the infection after being treated for rejection. However, in a series of over 500 recipients of transplants for HCV in our institution, spontaneous HCV clearance was not observed. Despite these limitations, our experience would argue that interferon therapy is not without risks. Because the exact mechanism of action of interferon in inducing liver allograft rejection remains obscure, prospective trials are needed in order to better assess the safety of such therapy in an immunosuppressed population.
 
In conclusion, with recurrent HCV becoming increasingly common in the transplant community, there will be a greater need to find safe and effective antiviral therapy. The risk of acute rejection and subsequent graft loss and the risk of accelerated recurrent HCV with subsequent cirrhosis in LT patients on interferon should be weighed heavily in choosing treatment for recurrent HCV, which itself can cause allograft failure if untreated. Treating physicians should therefore be aware of the risks and monitor patients accordingly.
 
 
 
 
 
  icon paper stack View Older Articles   Back to Top   www.natap.org