Anti-GBM (Goodpasture) disease: Treatment and prognosis

INTRODUCTION — Anti-glomerular basement membrane (anti-GBM) disease is a rare small vessel vasculitis that affects glomerular capillaries, pulmonary capillaries, or both. Most patients present with rapidly progressive (crescentic) glomerulonephritis, although some patients may present with relatively mild kidney impairment. In general, this disorder is typically associated with severe kidney injury that, if untreated, progresses quickly to end-stage kidney disease (ESKD).

The treatment and prognosis of anti-GBM disease are discussed in this review. The pathogenesis, clinical manifestations, and diagnosis of this disorder are discussed elsewhere. (See "Anti-GBM (Goodpasture) disease: Pathogenesis, clinical manifestations, and diagnosis".)

IMPORTANCE OF EARLY INTERVENTION — Early diagnosis and intervention are key determinants of the response to therapy and long-term prognosis of patients with anti-GBM disease [1]. There is a direct correlation between the initial plasma creatinine concentration and the percent of glomeruli with crescents; in particular, crescents are usually present in more than 75 percent of glomeruli when the plasma creatinine concentration is above 5 mg/dL (442 micromol/L). In addition, there is a direct correlation between anti-GBM antibody levels and the plasma creatinine at presentation [2]. Avoidance of maintenance dialysis is uncommon in patients who require dialysis within 72 hours of presentation, particularly in those who have crescents involving all glomeruli [3]. By comparison, prevention of end-stage kidney disease (ESKD) can usually be achieved in less severe cases, although some do progress. A low proportion of preserved glomeruli and the presence of oligoanuria may be the best determinants of poor prognosis for recovery. (See 'Prognosis' below.)

A common clinical question is whether to start empiric therapy in patients with suspected anti-GBM disease before the diagnosis has been confirmed by serologic testing or kidney biopsy. We do not delay therapy in these patients and initiate therapy within 24 hours of a presumptive diagnosis. The results of diagnostic testing may take several days to become available, and delaying treatment may increase the patient's risk of developing irreversible kidney damage.

WHOM TO TREAT — In most patients with anti-GBM disease, we recommend treatment with plasmapheresis combined with prednisone and cyclophosphamide (see 'Initial therapy' below). This includes the following:

●All patients with pulmonary hemorrhage (hemoptysis), independent of the presence and/or severity of kidney involvement.

●All patients with kidney involvement who do not require immediate dialysis.

Whether to treat patients who present with dialysis-dependent kidney failure without pulmonary hemorrhage is a more challenging decision. Some experts do not treat these patients, since there is a very low likelihood of kidney response, especially if the patient has 100 percent crescents on kidney biopsy [3]. Other experts consider a short trial (two to three weeks) of plasmapheresis and immunosuppressive therapy, particularly among the following patients:

●Patients with very acute disease, in whom irreversible injury is less predictable [4].

●Younger patients who are better able to tolerate aggressive immunosuppression.

●Patients whose biopsy shows focal crescentic glomerular damage associated with acute tubular injury [3].

●Patients with anti-GBM disease who have both antineutrophil cytoplasmic autoantibodies (ANCA) and clinical signs of a systemic vasculitis [5-9]. As an example, findings such as a purpuric rash and arthralgias are suggestive of a concurrent vasculitis since they are uncommon in anti-GBM disease alone. (See 'Double-positive anti-GBM and ANCA-associated disease' below and "Granulomatosis with polyangiitis and microscopic polyangiitis: Clinical manifestations and diagnosis".)

Our rationale to treat all patients with kidney involvement who do not require immediate dialysis is based upon observational evidence that immunosuppressive therapy may also benefit patients who present with high plasma creatinine concentrations. In a retrospective analysis of 71 patients with anti-GBM disease who were treated with plasma exchange, prednisolone, and cyclophosphamide, patient and kidney survival varied with the severity at presentation:

●For patients who presented with a plasma creatinine concentration of less than 5.7 mg/dL (500 micromol/L), patient and kidney survival were 100 and 95 percent at one year and 84 and 74 percent at last follow-up (median period of 90 months), respectively.

●For patients who presented with a plasma creatinine concentration greater than 5.7 mg/dL (500 micromol/L), but who did not require immediate dialysis (within 72 hours of presentation), patient and kidney survival were 83 and 82 percent at one year and 72 and 69 percent at last follow-up, respectively.

●For patients who required immediate dialysis, patient and kidney survival were 65 and 8 percent at one year and 36 and 5 percent at last follow-up, respectively. All such patients who had crescents in all glomeruli on kidney biopsy required long-term maintenance dialysis. By comparison, two patients with significant acute tubular necrosis plus crescents on biopsy eventually recovered long-term independent kidney function.

●Among the 42 patients with pulmonary hemorrhage, bleeding resolved in approximately 90 percent.

INITIAL THERAPY

Plasmapheresis plus immunosuppressive therapy — Once a decision is made to treat (see 'Whom to treat' above), we recommend treatment with plasmapheresis combined with immunosuppressive therapy, rather than immunosuppressive therapy alone. Plasmapheresis removes circulating anti-GBM antibodies and other mediators of inflammation, and the immunosuppressive agents minimize new antibody formation. In most cases, we advocate for inpatient admission to initiate therapy.

There are no large, randomized trials to support this approach, and most studies have been uncontrolled. In general, the available data suggest that approximately 30 to 45 percent of patients treated with plasmapheresis combined with immunosuppression will benefit by not progressing to end-stage kidney disease (ESKD) or death [5,10-13]. However, recovery is much more likely in patients who begin treatment before oligoanuria ensues and is uncommon in patients who require dialysis or who have 100 percent crescents on biopsy. (See 'Whom to treat' above.)

The only available randomized trial evaluated outcomes among 17 patients who were treated with prednisone and cyclophosphamide with or without plasmapheresis [14]. At the end of treatment, two of eight patients in the plasmapheresis group, compared with six of nine in the immunosuppression-alone group, became dialysis dependent (25 versus 67 percent, respectively). The authors concluded that, although there may have been some benefit from plasmapheresis, the percent of crescents on initial kidney biopsy and entry plasma creatinine correlated better with outcome. Regardless of therapy, patients with less than 30 percent crescents and a plasma creatinine below 3 mg/dL (265 micromol/L) did well, while those with severe crescentic involvement and a plasma creatinine above 4 mg/dL (354 micromol/L) did poorly. In observational studies, the use of plasmapheresis has been associated with improved patient and kidney survival [15,16].

Despite the absence of definitive evidence of benefit, plasmapheresis is generally recommended for the treatment of patients with anti-GBM disease. Two factors are considered by many experts to justify this recommendation:

●Improved morbidity and mortality in the era of plasmapheresis compared with historic rates

●Biological plausibility of greater amelioration of the consequences of disease with rapid removal of anti-GBM antibody, compared with a slower reduction in levels seen with immunosuppressive agents alone

Plasmapheresis regimen — Our preferred initial plasmapheresis prescription is daily 4 liter exchanges for two to three weeks [2,3,17]. However, alternate day plasmapheresis has been used by some, especially after the initial two weeks of treatment. Plasmapheresis using either centrifugal or membrane separation methods is considered to be equally efficacious [18]. In general, albumin is given as the replacement fluid. If, however, the patient has had a recent kidney biopsy or has pulmonary hemorrhage, then one to two liters of fresh frozen plasma should be substituted for albumin at the end of the procedure to replete coagulation factors that may have been depleted by plasmapheresis [19].

At the end of this two- to three-week regimen, the need for further plasmapheresis is determined by the patient's clinical status and serum anti-GBM antibody titers (see 'Monitoring the response to therapy' below). Further plasmapheresis may be unnecessary if the patient has improved and serum anti-GBM antibody titers have markedly declined and are undetectable or nearly undetectable. By comparison, continued plasmapheresis may be required if the patient still has active pulmonary disease (eg, hemoptysis), if antibody titers are not declining substantially, or if antibody titers rebound after stopping plasmapheresis [2].

A potential complication with fresh frozen plasma (14 percent citrate by volume) is the development of metabolic alkalosis. Metabolism of the administered citrate generates bicarbonate, the excretion of which may be limited by kidney failure. Citrate-induced metabolic alkalosis can be corrected by hemodialysis if needed. (See "Therapeutic apheresis (plasma exchange or cytapheresis): Complications", section on 'Citrate-induced metabolic alkalosis'.)

If there is a severe infection in the setting of plasmapheresis, a single infusion of intravenous immune globulin (IVIG; 100 to 400 mg/kg) can be given after a plasmapheresis session to partially replenish immunoglobulin levels [20].

Immunosuppressive therapy

Glucocorticoids plus cyclophosphamide — Plasmapheresis must be accompanied by glucocorticoids and immunosuppressive agents, preferably cyclophosphamide.

●Glucocorticoids – We administer pulse intravenous (IV) methylprednisolone (15 to 30 mg/kg to a maximum dose of 1000 mg over 20 minutes) daily for three doses followed by daily oral prednisone (1 mg/kg per day to a maximum of 60 to 80 mg/day), which can be tapered once remission is induced. However, there is no evidence of benefit of pulse methylprednisolone over other glucocorticoid doses and schedules, and some clinicians treat with oral glucocorticoids without an IV pulse [3]. On plasmapheresis days, we administer prednisone after plasmapheresis, although this agent is minimally removed by plasmapheresis [21]. Glucocorticoids are generally continued for up to six months.

●Cyclophosphamide – The initial cyclophosphamide dose is 2 mg/kg per day orally. We typically reduce the dose by 25 percent in older adults (age >60 years) and frail patients and adjust the dose appropriately for impaired kidney function (table 1). On plasmapheresis days, we administer cyclophosphamide after plasmapheresis. We continue cyclophosphamide for three months in most patients; if anti-GBM antibody levels are not substantially reduced by three months (see 'Monitoring the response to therapy' below), we continue cyclophosphamide for up to a maximum of six months.

In patients who cannot take oral medications, IV therapy may be used. The optimal dosing of IV cyclophosphamide is uncertain. Some centers use IV protocols similar to those used for the treatment of antineutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis or lupus nephritis; however, there is no evidence to support the use of these regimens in patients with anti-GBM disease. (See "Granulomatosis with polyangiitis and microscopic polyangiitis: Induction and maintenance therapy", section on 'Cyclophosphamide-based regimen' and "Lupus nephritis: Initial and subsequent therapy for focal or diffuse lupus nephritis", section on 'Cyclophosphamide-based regimen'.)

Most patients with anti-GBM disease who experience remission do not require maintenance immunosuppressive therapy given the low rate of recurrent disease. An exception to this is the patient who is double-positive for anti-GBM antibodies and ANCA. Such patients have a higher risk of relapse of vasculitis and should receive maintenance therapy similar to that used for ANCA-associated vasculitis. (See 'Patients with recurrent disease' below and 'Double-positive anti-GBM and ANCA-associated disease' below.)

Although we generally use oral cyclophosphamide, the relative efficacy of oral and IV cyclophosphamide in anti-GBM disease is not known. However, in a retrospective study of 122 patients with anti-GBM disease treated with plasma exchange, of whom 32 received oral and 71 received IV cyclophosphamide, there was a nonsignificant trend toward improved patient survival among those receiving oral cyclophosphamide [22]. IV therapy may be used in patients who cannot take oral medications.

Cyclophosphamide (oral or IV) can be associated with Pneumocystic jirovecii (carinii) pneumonia, amenorrhea, alopecia, and bladder toxicity (hemorrhagic cystitis and bladder cancer). Complications with high doses of glucocorticoids include oropharyngeal fungal infections, gastritis (which can result in gastrointestinal bleeding in patients at increased risk), and bone loss. Both cyclophosphamide and glucocorticoids are associated with an increased risk of infection. These complications and recommended prophylactic measures are discussed in detail elsewhere:

●(See "Epidemiology, clinical manifestations, and diagnosis of Pneumocystis pneumonia in patients without HIV".)

●(See "General principles of the use of cyclophosphamide in rheumatic diseases".)

●(See "Prevention and treatment of glucocorticoid-induced osteoporosis".)

Alternatives to cyclophosphamide — Rituximab and mycophenolate mofetil (MMF) are alternative therapies that can be used in patients who are unable to tolerate cyclophosphamide. However, we do not usually use these agents as first-line therapy because of insufficient evidence for their efficacy:

●Rituximab – If rituximab is used, we give 1 g for two doses. If the patient is receiving daily plasmapheresis, the first of the two rituximab doses can be given after the initial seven consecutive days of plasmapheresis and glucocorticoids, since concurrent rituximab and plasmapheresis will result in removal of rituximab from the circulation. After a 48-hour period has elapsed, another seven days of plasmapheresis can be performed, after which the second of the two doses of rituximab is given. If the patient is receiving alternate-day plasmapheresis, rituximab should be given immediately after the exchange, which will permit time for binding of rituximab to B cells prior to the next plasmapheresis treatment.

●Mycophenolate mofetil – If mycophenolate mofetil (MMF) is used, we administer an initial dose of 500 mg twice daily and titrate up as tolerated to a dose of 1000 mg twice daily. We monitor the white blood cell count for signs of leukopenia. Dose reduction is warranted in patients with leukopenia or significant gastrointestinal side effects.

Several reported cases have documented successful treatment of anti-GBM disease with these agents [23-27]; however, evidence in the form of controlled trials is lacking. In a review of 22 patients treated with rituximab, 15 were treated for relapsed or refractory disease, and seven received the drug as initial therapy (instead of cyclophosphamide) [28]. No patients had concurrent ANCA positivity. Pulmonary hemorrhage resolved and anti-GBM titers decreased in nearly all patients, but none of the patients with refractory disease recovered kidney function.

Supportive measures for severe disease — In addition to plasmapheresis and immunosuppressive therapy, patients with severe disease may require organ supportive care. Patients presenting with life-threatening hemoptysis, for example, may require intubation and mechanical ventilation as well as measures to control the bleeding. These issues are discussed in greater detail elsewhere. (See "Evaluation and management of life-threatening hemoptysis".)

Patients who develop severe kidney failure in spite of treatment may require initiation of dialysis. The indications for and timing of kidney replacement therapy are the same as those for other causes of kidney failure and are discussed separately. (See "Kidney replacement therapy (dialysis) in acute kidney injury in adults: Indications, timing, and dialysis dose".)

Monitoring the response to therapy — All patients with anti-GBM disease should be closely monitored during treatment. Our approach is as follows:

●Clinical status – During the acute phase of the disease (first two to three weeks), when most patients are admitted to the hospital for treatment, we monitor the serum creatinine level, a complete blood count, and urine output daily. If the patient has pulmonary hemorrhage, serial chest radiographs are reasonable to monitor for worsening alveolar hemorrhage. When the patient's condition is stable enough for discharge from the hospital, we typically schedule follow-up visits weekly for the first two to four weeks. In patients who continue to remain stable, the duration between follow-up visits can then be extended to every two to four weeks.

●Anti-GBM antibody levels – We monitor anti-GBM antibody levels weekly for the first six weeks until they are undetectable on two consecutive occasions. We subsequently monitor anti-GBM antibody levels every other week for four weeks, and if they remain persistently undetectable, we monitor levels once monthly for six months. In addition, we check antibody levels if the patient experiences clinical signs suggestive of recurrence (see 'Patients with recurrent disease' below). In patients receiving plasmapheresis combined with glucocorticoids and cyclophosphamide, disappearance of anti-GBM antibodies typically occurs within four weeks of treatment; the continued presence of antibodies beyond eight weeks is uncommon [2]. If antibody levels remain persistently elevated, the immunosuppressive regimen may require modification. (See 'Patients with persistent anti-GBM antibodies after initial therapy' below.)

PATIENTS WITH RECURRENT DISEASE — Relapses are uncommon (around 2 percent in one center's experience), but data are not sufficient to reliably determine how often this occurs [3,29-32]. Clinical relapses are more common in patients who are also antineutrophil cytoplasmic autoantibody (ANCA)-positive, in whom it is the vasculitis and not the anti-GBM disease that is reactivated [33]. (See 'Double-positive anti-GBM and ANCA-associated disease' below.)

There may be a higher rate of recurrence in patients who are smokers or have exposure to hydrocarbon in their occupation. All patients with anti-GBM disease should refrain from smoking and avoid such exposures.

In cases of potential recurrence with kidney involvement, we repeat a kidney biopsy to confirm the diagnosis and exclude concomitant pathologies such as ANCA-associated vasculitis and membranous nephropathy. In confirmed cases of recurrent anti-GBM disease, we retreat with plasmapheresis combined with glucocorticoids and cyclophosphamide using the same regimen as described above for initial therapy. (See 'Plasmapheresis plus immunosuppressive therapy' above.)

Rituximab or mycophenolate mofetil (MMF) may be considered as alternative therapies to cyclophosphamide in patients who develop recurrent disease or relapse while on cyclophosphamide or are unable to tolerate this drug. (See 'Alternatives to cyclophosphamide' above.)

PATIENTS WITH PERSISTENT ANTI-GBM ANTIBODIES AFTER INITIAL THERAPY — Uncommonly, patients with anti-GBM disease have persistently elevated anti-GBM antibody levels, with or without active disease, in spite of treatment with daily plasmapheresis and immunosuppressive therapy for at least two weeks. The optimal management of such patients is not known, and data are limited to case reports and small case series [27,28]. Not all patients with residual anti-GBM antibody titers will require a change in therapy, however, and treatment should be individualized based upon the patient's clinical picture. The following clinical scenarios illustrate the range of decision making:

●In patients with a persistent high antibody titer (arbitrarily defined as three times the upper limit of normal or higher) that is not decreasing in spite of daily plasmapheresis with cyclophosphamide-based therapy for at least two to three weeks and who have evidence of ongoing disease activity (ie, ongoing pulmonary hemorrhage and/or active glomerulonephritis [persistent or new red blood cell casts in the urine and/or worsening kidney function]), some experts give rituximab (1 g initially followed 14 days later by another 1 g dose). The need for further daily plasmapheresis beyond four weeks should be reassessed, particularly in patients who have not yet recovered kidney function by one month of initial therapy and do not have pulmonary hemorrhage. (See 'Alternatives to cyclophosphamide' above.)

●In patients with a persistent high antibody titer who are on dialysis, have no pulmonary hemorrhage, and have a kidney biopsy showing a high percentage of crescents or glomerulosclerosis and interstitial fibrosis, we taper off treatment and continue to monitor anti-GBM antibody levels.

●In patients with a persistent intermediate antibody titer (arbitrarily defined as one to three times the upper limit of normal) after three to four months of cyclophosphamide-based therapy who have recovered kidney function and are overall doing well, we switch from cyclophosphamide to rituximab (1 g initially followed 14 days later by another 1 g dose). If rituximab is not available, we switch to azathioprine (1 to 2 mg/kg per day) or mycophenolate mofetil (MMF; 1000 mg twice daily), both of which are less toxic than cyclophosphamide, and continue treatment for six to nine months. We continue to monitor such patients for signs of clinical recurrence or rising anti-GBM antibodies. (See 'Alternatives to cyclophosphamide' above.)

SPECIAL POPULATIONS

Double-positive anti-GBM and ANCA-associated disease — Patients who are double positive for anti-GBM antibodies and antineutrophil cytoplasmic autoantibodies (ANCA) should be managed initially as for anti-GBM disease since that is the more severe lesion. However, unlike patients with single-positive anti-GBM disease, double-positive patients will require maintenance immunosuppressive therapy for ANCA disease because of the tendency of vasculitis to relapse [33]. Because of reports that double-positive patients may be more likely to recover from dialysis than single-positive anti-GBM disease, treatment with plasmapheresis plus immunosuppressive therapy should be considered for those presenting with dialysis-requiring kidney failure [33]. (See "Granulomatosis with polyangiitis and microscopic polyangiitis: Induction and maintenance therapy", section on 'Maintenance therapy'.)

Observational studies have shown that double-positive patients have similar outcomes to those with single-positive anti-GBM disease but experience higher rates of relapse because they have ANCA-associated vasculitis [33,34]. As an example, one study that included 41 patients with single-positive anti-GBM disease and 37 double-positive patients reported 12-month kidney survival rates of 44 and 53 percent, respectively [33]. Kidney recovery at one year was greater among double-positive patients than among those with anti-GBM disease (29 versus 17 percent). However, while no single-positive anti-GBM disease patients experienced disease relapse, approximately one-half of double-positive patients had recurrent disease over a median of 4.8 years, suggesting that maintenance immunosuppressive therapy may be beneficial in these patients, as in those with ANCA-associated vasculitis.

Anti-GBM disease associated with membranous nephropathy — Several series of patients with anti-GBM disease associated with membranous nephropathy have been described [35]. We treat such patients using the same approach as that used for patients with anti-GBM disease (see 'Initial therapy' above). Treatment response appears to be similar to that of patients with anti-GBM disease without membranous nephropathy, and, in some cases, proteinuria attributed to membranous nephropathy has also improved [36]. (See "Anti-GBM (Goodpasture) disease: Pathogenesis, clinical manifestations, and diagnosis", section on 'Anti-GBM disease associated with membranous nephropathy'.)

Anti-GBM disease without detectable circulating anti-GBM antibodies — There are infrequent reports of patients with positive linear staining for immunoglobulin G (IgG) on kidney biopsy and negative commercial testing for circulating anti-GBM antibodies. (See "Anti-GBM (Goodpasture) disease: Pathogenesis, clinical manifestations, and diagnosis", section on 'Anti-GBM disease without detectable circulating anti-GBM antibodies'.)

Some of these patients without detectable circulating anti-GBM antibodies may have a variant known as "atypical anti-GBM nephritis," which is an indolent form of anti-GBM disease without pulmonary involvement [37]. Diffuse crescentic disease is uncommon in these patients. In such patients without significant crescentic glomerulonephritis, we do not administer treatment with immunosuppressive agents. However, in rare patients who develop signs of progressive disease, we treat with glucocorticoids plus cyclophosphamide without plasmapheresis since there are no detectable circulating antibodies to remove with plasmapheresis. (See 'Glucocorticoids plus cyclophosphamide' above.)

A subset of patients with biopsy-proven anti-GBM disease may have undetectable circulating anti-GBM antibodies due to false-negative testing. If the suspicion for anti-GBM disease is high in such patients based upon clinical presentation, we treat using the same approach as that used for patients with anti-GBM disease who have detectable anti-GBM antibodies. Alternative methods of testing for circulating anti-GBM antibodies, such as indirect immunofluorescence, Western blot, or biosensor system, should be performed if available. If repeat testing using these methods is also negative, we discontinue plasmapheresis and continue treatment with glucocorticoids plus cyclophosphamide. (See "Anti-GBM (Goodpasture) disease: Pathogenesis, clinical manifestations, and diagnosis", section on 'Anti-GBM disease without detectable circulating anti-GBM antibodies'.)

Recurrent anti-GBM disease after transplantation — Recurrence of anti-GBM disease after kidney transplantation is rare. Treatment of this disorder is discussed elsewhere. (See "Anti-GBM (Goodpasture) disease: Recurrence after transplantation".)

INVESTIGATIONAL THERAPIES — A number of investigational therapies have been proposed for the treatment of anti-GBM disease:

●Immunoadsorption – Immunoadsorption, performed as part of the plasmapheresis procedure, may be effective in some patients with anti-GBM disease, even in those who are dialysis dependent. In one such individual, for example, the combined use of immunosuppression and immunoadsorption using a sepharose-coupled, sheep-antihuman IgG column for 25 cycles resulted in the recovery of kidney function, with a stable creatinine concentration of 2 mg/dL (177 micromol/L) [38]. Other small observational studies have also shown possible benefits of immunoadsorption [39,40].

●Imlifidase – The IgG-degrading enzyme derived from Streptococcus pyogenes (imlifidase [IdeS]) is a recombinant cysteine protease that cleaves all four subclasses of human IgG into F(ab’)₂ and Fc fragments, inhibiting complement-dependent and antibody-dependent cellular cytotoxicity. The efficacy and safety of this agent were evaluated in a phase 2a single-arm study of 15 patients with circulating anti-GBM antibodies and an eGFR <15 mL/min/1.73 m² who received a single dose of imlifidase (0.25 mg/kg) [41]. All patients also received standard therapy with cyclophosphamide and glucocorticoids; plasmapheresis was administered only if autoantibodies rebounded. At baseline, 10 patients were dialysis dependent and five had an eGFR between 7 and 14 mL/min/1.73 m²; six patients were also positive for antineutrophil cytoplasmic autoantibodies (ANCA). At six months, 10 of 15 patients (67 percent) were dialysis independent, which was a better outcome when compared with a historical control cohort (18 percent). Eight serious adverse events were reported (including one death), although none were assessed as probably or possibly related to therapy. Additional studies are required to confirm these findings.

KIDNEY TRANSPLANTATION — Patients who develop end-stage kidney disease (ESKD) as a result of anti-GBM disease may be candidates for kidney transplantation. There are no data to guide the optimal timing of transplantation in this setting. Most transplant centers require at least six months of undetectable anti-GBM antibody levels before kidney transplantation.

Data on patient outcomes of after kidney transplantation are limited:

●In a multicenter retrospective cohort study of 53 patients with anti-GBM disease who underwent kidney transplantation from 1977 to 2015, recurrence occurred in a first kidney transplant in only one patient five years posttransplant in the context of cessation of immunosuppressive medications [42]. Patient survival was 100, 94, and 89 percent at 5, 10, and 15 years, respectively. Death-censored first-graft survival rates were 88, 83, and 79 percent at 5, 10, and 15 years, respectively.

●In a single-center study of 26 patients with anti-GBM disease who underwent kidney transplantation between 1994 and 2015, both graft loss and patient mortality were low and similar to that of patients with IgA nephropathy who underwent transplantation [43]. Disease recurrence occurred in only one patient.

Recurrence of anti-GBM disease after kidney transplantation is discussed in more detail elsewhere. (See "Anti-GBM (Goodpasture) disease: Recurrence after transplantation".)

PROGNOSIS — In view of the self-limited nature of anti-GBM disease, patients who survive the first year with intact kidney function generally do well. As previously discussed, kidney and patient survival correlates closely with the degree of kidney impairment at presentation [3,44,45]. Patients with moderate to severe disease who do not require dialysis upon presentation generally respond to therapy, with recovery being maintained during long-term follow-up. By comparison, few who require immediate dialysis escape the need for maintenance dialysis.

Other clinical and pathologic features have also been shown to predict kidney outcomes among patients with anti-GBM disease:

●In a multicenter series of 123 patients diagnosed with anti-GBM disease between 1986 and 2015, five-year kidney and patient survival rates were 34 and 83 percent, respectively [44]. In a multivariable analysis, dialysis dependency at presentation was the strongest predictor of end-stage kidney disease (ESKD) during follow-up (hazard ratio [HR] 3.17, 95% CI 1.59-6.32). In addition, certain histopathologic features, such as the percentage of normal glomeruli and the extent of interstitial infiltrate on kidney biopsy, were also identified as independent predictors of ESKD.

●In another multicenter series of 119 patients with anti-GBM disease, of whom 78 percent required dialysis at presentation and 46 percent had pulmonary hemorrhage, five-year patient survival was 92 percent [16]. Risk factors for death included age at onset (HR 4.10 per decade, 95% CI 1.89-8.88), hypertension (HR 19.9, 95% CI 2.52-157), dyslipidemia (HR 11.1, 95% CI 2.72-45), and need for mechanical ventilation (HR 5.20, 95% CI 1.02-26.4). The use of plasmapheresis was associated with improved survival (HR 0.29, 95% CI 0.08-0.98). Kidney survival at three months was 31 percent. Among the patients who required dialysis at presentation, only 16 percent recovered kidney function.

●In a study of 43 patients with anti-GBM disease, oligoanuria at diagnosis was the strongest predictor of patient mortality, and both oligoanuria and the percentage of glomerular crescents seen on biopsy were the best predictors of a poor kidney outcome [34].

●A multicenter, international cohort study of 174 patients with newly diagnosed anti-GBM disease found that a renal risk score (RRS), used to predict kidney outcomes in patients with ANCA-associated vasculitis, was a strong predictor for ESKD [46]. The RRS for each patient was calculated using eGFR at presentation, percentage of normal glomeruli in the kidney biopsy, and the simplified cutoff for tubular atrophy and interstitial fibrosis. Kidney survival at 36 months was 100, 62, and 21 percent in the low-, moderate-, and high-risk groups, respectively. The need for kidney replacement therapy (KRT) at diagnosis and the percentage of normal glomeruli in the kidney biopsy were also found to be independent predictors of ESKD.

The outcome in patients with recurrent disease, whether antineutrophil cytoplasmic autoantibody (ANCA) positive or negative, is typically superior to that in the initial presentation of anti-GBM disease [47]. In this setting, the diagnosis is usually clear, leading to the quick initiation of appropriate therapy. (See 'Patients with recurrent disease' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Glomerular disease in adults".)

SUMMARY AND RECOMMENDATIONS

●Importance of early intervention – If left untreated, anti-glomerular basement membrane (anti-GBM) disease usually progresses rapidly to end-stage kidney disease (ESKD). Early diagnosis and intervention are critical for achieving the best response to therapy since the percent of crescents on initial kidney biopsy and the pretherapy plasma creatinine concentration correlate best with outcome. Recovery of kidney function is rare if dialysis is required at initiation of treatment. (See 'Introduction' above and 'Importance of early intervention' above.)

●Initial therapy – For most patients with anti-GBM disease, we recommend plasmapheresis plus immunosuppressive therapy, rather than immunosuppressive therapy alone (Grade 1C). This includes all patients with pulmonary hemorrhage (hemoptysis), independent of the presence and/or severity of kidney involvement, and all patients with kidney involvement who do not require immediate dialysis. In selected patients who require immediate dialysis and do not have pulmonary hemorrhage, such as those with very acute disease, those with biopsies showing focal crescents and tubular damage, or those with antineutrophil cytoplasmic antibodies (ANCA) and clinical signs of systemic vasculitis, a trial of therapy can be considered. (See 'Whom to treat' above.)

•Plasmapheresis – We perform daily 4 liter exchanges for two to three weeks. In general, albumin is given as the replacement fluid. If, however, the patient has had a recent kidney biopsy or has pulmonary hemorrhage, then one to two liters of fresh frozen plasma should be substituted for albumin at the end of the procedure to replete coagulation factors that may have been depleted by plasmapheresis. (See 'Plasmapheresis regimen' above.)

•Immunosuppression – We suggest the combination of glucocorticoids plus cyclophosphamide rather than other immunosuppressive agents (Grade 2C). We administer pulse methylprednisolone (15 to 30 mg/kg to a maximum dose of 1000 mg intravenously [IV] over 20 minutes) daily for three doses followed by daily oral prednisone (1 mg/kg per day to a maximum of 60 to 80 mg/day), which can be tapered once remission is induced. Some clinicians choose to administer oral prednisone without an IV pulse. The initial cyclophosphamide dose is 2 mg/kg per day orally. We typically reduce the dose by 25 percent in older adults (age >60 years) and frail patients and adjust the dose appropriately for impaired kidney function (table 1). (See 'Glucocorticoids plus cyclophosphamide' above.)

Patients who either refuse or, because of severe side effects, need to discontinue cyclophosphamide may receive rituximab therapy, or alternatively, mycophenolate mofetil (MMF). Given the insufficient evidence for these agents, however, we do not advocate their use as first-line therapy for anti-GBM disease. (See 'Alternatives to cyclophosphamide' above.)

•Supportive measures – In addition to plasmapheresis and immunosuppressive therapy, patients with severe disease may require organ supportive care. Patients presenting with life-threatening hemoptysis, for example, may require intubation and mechanical ventilation as well as measures to control the bleeding. (See "Evaluation and management of life-threatening hemoptysis".)

●Monitoring – Anti-GBM antibody levels should be monitored weekly for the first six weeks until they are undetectable on two consecutive occasions. We subsequently monitor anti-GBM antibody levels every other week for four weeks, and if they remain persistently undetectable, we monitor levels once monthly for six months. In addition, we check antibody levels if the patient experiences clinical signs suggestive of recurrence (see 'Monitoring the response to therapy' above).

●Recurrent disease – Relapses are uncommon. In cases of recurrence with kidney involvement, we repeat a kidney biopsy to confirm the diagnosis and exclude concomitant pathologies such as ANCA-associated vasculitis and membranous nephropathy. In confirmed cases of recurrent anti-GBM disease, we retreat with plasmapheresis combined with glucocorticoids and cyclophosphamide using the same regimen as described above for initial therapy. (See 'Patients with recurrent disease' above.)