Saturday, February 23, 2008

HERMES trial for relapsing multiple sclerosis NEJM article

Hauser S SL, Waubant W. Arnold DL, Vollmer T, et al for HERMES Trial Group. B Cell Depletion with rituximab in relapsing remitting multiple sclerosis. NEJM 2008; 358: 676-678

and editorial
McFarland HF. Focus on research: the B Cell -- old player, new position on the team. NEJM 2008; 358: 664-666.

The phase 2 trial of Rituxan (Rituximab, Biogen Idex and Genentech) in relapsing remitting multiple sclerosis showed a significant reduction of relapses and of MRI lesions, the primary outcome measure. It was a short trial (48 weeks) and not powered to show an effect on disability. There was a rapid reduction not only of contrast enhancing lesions on MRI but also total lesions, within 4 weeks. McFarland opines that the benefits of led to increased interest in the role of B cells in diseases thought to be mediated by T cells. Some patients may have antibody mediated complement destructive lesions and in others B cells may precede lesion development. However the rapidity of the response suggests another mechanism. Possibly, B cells are involved in antigen presentation to T cells and activation of T cell programming (cites Rock, 1984). They may contribute to T cell priming and lesion formation. Treated patients showed the subsequent development of naive rather than memory B cells.

Many trials have shown a reduction of MRI lesions but not clinical disease activity. Since antibodies maybe involved in repair mechanisms, clinical status of the patients will need to be closely monitored in the phase 3 trial. B cells may alos be important within lesions, as CXCL13 which regulated B cell migration in lymphoid tissue is found within plaques.

The study is a phase 2 trial, which means that unanswered questions include low frequency side effects, duration of oeffect, and effect on disability. The study involves only 104 patients.

Monday, February 11, 2008

Side effects of interferon beta

1. cutaneous reaction
2. hyperthyroidism (Schwid et al. Arch Neurol 1997;54:1165-1190)
3. transvaginal bleeding (Pakulski et al., 1997 Ann Pharmacother311:50-52)/
4. liver carcinoma (Makita et al., Nippon Shokakdyo, Sasshi 1996;93:406-410)
5. psoriasis (Webster et al., J Am Acad Derm 1996; 34:365-367).
6. transitory hearing loss (Kanda et al., Audiology 1995; 34: 98-102)
7. polymyositis
8. intraocular bleeding
9. Severe anaphylaxis-- sudden laryngospasm and shock after taking drug for six months
10. flulike illness, fatigue and headache
11. depression

Sunday, February 03, 2008

Consensus statement on CSF standards in diagnosis of MS

Freedman MS, Thompson EJ, Deisenhammer F, et al. Recommended standard of cerebrospinal fluid analysis in the diagnosis of multiple sclerosis. A Consensus statement. Arch Neurol 2005; 62:865-870.

general points
1. The IgG index or any other quantitative IgG anaysis is not equivalent to qualitative analysis using isoelectric focusing with immunofixation, as opposed to the previous recommendation that equated the IgG index with qualitative analysis.

There is complete agreement that isoelectric focusing (IEF) on agarose gels followed by immunoblotting should be the gold standard for detecting oligoclonal bands. Other methods such as polyacrylamide gels combined with IEF and silver staining of proteins might have proved useful IN THE PAST but they LACK SPECIFICITY for IgG and are not supported by consensus. Direct silver staining techniques demonstrate reduced sensitivity and specificity.

Some techniques also stain kappa and gamma light chains (both free and bound) that discern faint bands better against polyclonal background. Light chain staining will also be positive in rare cases where OCB's are caused by the presence of IgA or IgM which will not appear on gels stained only for IgG. These add little to routine MS diagnosis.

Serum free chains are usually removed by kidney (free chains). Intrathecal IgG is usuallyy associated with kappa chains.

Sensitivity of using IEF with immunoblotting is in excess of 95 %. A negative test is an indication to rethink the diagnosis. The presence of only one band in CSF not in serum may indicate the need to retest as many patients will convert. There is a high negative predictive power of negative CSF in clinically isolated syndrome (CIS) such that a negative CSF exam indicates a low likelihood of developing MS. There are five patterns accepted (1- no bands in CSF or serum 2- OCB in CSF not in serum 3-- OCBs in CSF, identical bands in serum (still there is CSF IgG synthesis) 4-- same as 3 except with leaky BBB suggesting systemic oligoclonal band production 5-- Monoclonal bands in CSF and sera, suggestive of a monoclonal IgG protein present).

Q albumen (albumen quotient) can be used to assess CSF leakiness. Controls should be run with each sample to make sure that OCB's in positive controls are not overdeveloped and negative controls are not underdeveloped.

Patients deserve to have CSF analyzed at a lab that utilizes accepted standards.

Hematopoeitic stem cell transplantation for multiple sclerosis

Burt RK, Cohen B, Rose J et al. Arch Neurol 2005; 62: 860-864 Neurologic Review.

Authors comment intial attempts by hem-onc specialists used hem-onc protocols targeting SPMS patients. Recent trials from MS specialists with earlier patients are encouraging with less morbidity and mortality and improvement in EDSS. The key factors in success are selecting active patients (active by MRI or clinical criteria), treatment early before progressive disability ensues, and use of safer lymphoablative but not myeloablative HSCT conditioning regimen.

Rationale-- transplant ablates the aberrant disease causing immine cells while transplanted cells regenerate a new and antigen naive immune system.

Animal models -- Theiler murine encephalomyelitis virus (TMEV) in mice is a persistent viral related autoimmune disease, wherease EAE is a relapsing disease like MS. HSCT should be done while still in immune mediated inflammatory process not chronic progressive process.

Mobilizing hematopoietic stem cells (HSC's) from patients with MS-- HSCs are mobilized from peripheral blood from bone marrow with a growth factor such as granulocyte colony stimulating factor or chemotherapy (cyclophosphamide). GCF's can induce worsening of MS sometimes irreversibly and thus are prevented by administering steroids or cyclophosphamide. Ex vivo, lymphoctes and monocytes are purged and stem cells are selected, sometimes with monoclonal antibodies.

Conditioning regimen-- again the goal is lymphoablation not myeloablation. Myeloablative agents anyways will kill stem cells. Total body iradiation has the potential of causing neural stem cell apoptosis. Non myeloablative regiments include fludaribine, cyclophosphamide, Campath 1h, antithymocyte globulin. Fever induced pseudoexacerbations due to conduction blocks in marginally functional demyelinated axons should be avoided. The goals are 1) dose escale agents that work as conventional therapy 2) maximize immune suppression without myeloablation 3) avoid conditioning agents that injure disease affected and damaged CNS tissue 4) minimize the risk of fever 6) design justifiable regimens.

First generation protocols-- were extremely effective on MRI but not clinically due to late stage patients selected. tHERE WERE ALSO TREATMENT RELATED DEATHS.

Second generation therapies: the rationale for autologous HSCT is that MS is an environmentally caused disease, not a genetic stem cell defect. Nonmyeloablative treatments would presume to be more safe, following which HSC are infused to shorten period of cytopenias. Ruled OUT are etoposide, total body irradiation, busulfan, melphalan, and carmustine. Instead, cyclophosphamide and Campath 1H are used. Criteria are active disease despite treatment and EDSS 2.5-6.0, or higher if they have rapid deterioration and activity on MRI. A German protocol uses cyclophosphamide and rabbit antithymocyte globulin. At Northwestern, a regimen of cyclophosphamide and Campath 1H is used. More than half do not requrie a blood transfusion.

Comment-- to my knowledge this therapy is cash on the barrelhead, which has given some serious MS researchers pause in suggesting it.

MRI as a surrogate measure for disease activity in MS

See Neurology suppl. 3 2002 p. S29
Authors note that in the main clinical trials the correlations between relapse rate and EDSS progression with T2 burden are generallyin the range of 0.15 to 0.30. MRI cannot be used to predict EDSS progression or relapse rate. Autors cite the "Prentice criteria" a statistical look at what makes a surrogate marker. (Prentice R. Surrogate markers in clinicl trials: definition and operational criteria. Stat Med 1989; 8:431-440) states the surrogate marker must closely mirror the gold standard which here, is clinical activity.

Rating the main MS studies (the classics)

from Neurology suppl 59;S3 2002
CHAMPS Avonex v. Solumedrol/placebo Outcome point CDMS NOT disability, relapse, brain atrophy. NTT 6.7 to prevent one conversion to CDMS over 3 years. Class I Category A trial

Beta1a v. placebo. Primary endpoint was confimred EDSS progression between treatment group and controls. Relpase rate was a secondary endpoint. EDSS 1-3.5 at entry, 2 relapses in 3 years. Only 57 % of patients completed 2 years. Proportion with a one point increase in EDSS was 21.9 v. 34.9 % in treated/control groups, respectively. (p=0.02). Relapse rate improved by 18 %. Among the subgroup that completed 2 years of treatment there was a 31 % reduction of relapses. There was a 50 % reduction of Gad+ lesions but only 165/301 patients were included in this analysis. There wa no effect on T2 burden. NAbs were seen in 22 % by week 104. NTT to prevent one point increase in EDSS was 7.7. Class I Category A. However, it was short study was stopped early , studied a restricted population (mild entry EDSS 1-3.5) and showed decreased disability but no effect on T2 burden .

ETOMS Rebif 22 sq/per week v. placebo . Unlike CHAMPS, no steroids given in placebo group. No change in EDSS seen. MRI showed a benefit. NTT 9.1 to prevent one CDMS in 2 years. Class I Category A trial

PRISMS patients with 2 relapses in 2 years with EDSS 0-5.0 were reandomized to Rebif 22, Rebif 44 or placebo. Active treatment increased the time to progression of disability and reduced the probability of progression, and reduced the median integrated Disabiliyt status Scale score (area under the curve) mean increase in EDSS and mean increase in Ambulation Index. It reduced relapses, reduced moderate and severe relapses, number of steroid courses, and number of hospitalizations. It increased the number of relapse free patients. T2 burden increased by 10 % inplacebo and decreased with both treatment arms. Number of active lesions also was significantly reduced. 90 % completed 2 years and 95 % were followed for two years. NTT (high and low dose Rebif) to prevent one relapse per year, was 2.4, and 2.7, to render one patient relapse free for one year, 4.3 and 6.7; to render one patient relapse free for two years, 6.3 and 9.1; to prevent one moderate to severe relapse over two years, 2.7 and 3.6; and to prevent onepoint progression in EDSS, 9 and 12. Class I Category A. Rebif prevents progression of disability and reduces relapses in MS.

PRISMS 4-- extension of trial to four years. primary endpoint was relapse rate, with main comparison high dose angainst placebo. Original placebo patients were rerandomized to high or low dose. A dose effect relationship for disability was seen in years 3 and 4. Relapse rate was better with treatment but not with high dose treatment. The higher dose was significantly more effective for time to second relapse and the need for steroids. MRI was always better in the earlier treated patients. 20 % + of patients switching from placebo to drug developed NAbs. NTT not calculated. Class I, Category A. The trial established that Rebif 22 or 44 was effective in reducing relapses, delaying disability, improving MRI markers(t2 burden and no of new T2 lesions per scan) but no difference between high and low doses.

Interferon B1b in RRMS EDSS 0-5.5. gave betaseron 1.6 or 8 miu sq qod. Not well blinded, but randomized. No beneficial effect of EDSS found. Significant effect found on time to first relapse, proportion of patients relapse free, relapse rate, , number of moderate or severe relapses, and need for hospitalization over two years. MRI lesion burden increased by 20 % in placebo group, 10.5 % in low dose group, and decreased by 0.1 % in high dose group. NTT was 2.3 patients to prevent one relapse per year in this actively relapsing group. NTT 5.6 patients to keep one patient relapse free for two years, and 4.5 to prevent one moderate or severe relapse in one year. Class I category A

Extension study of Beta ib. Primary outcome measures were time to sustained worsening by one point on EDSS and mean change in each group from baseline. Not blinded. Study showed a trend (p=0.096) on time to sustained disability over five years with NO DATA GIVEN on other stated endpoint of mean change in EDSS. Both doses led to better relapse data for whole five year period. Rate of moderate and severe attacks was lower in both groups. Less patients completed years 3-5. MRI was beneficial for treated group although numbers were small at this point of study. There were 154 dropoutsby then en d of the study. NTT not calculated. Class II Category B, shows an effect for relapses only.

Copolymer (glatiramer acetate) GA v. placebo baseline EDSS of 0-5.5 . There was no effect of EDSS. Relapses were reduced 29 % over two years. MRI data not presented. NTT was 2.7 to prevent one relapse over two years. Class I Category A study for relapses.

Open label extension trial. Disability data gleaned against natural history data. Annual relapse rate of 0.42 over both phases of trial was stated as 72 % reduction over 2 years before study entry is misleading as it only was calculated with 83 patients and no account was made for regression to the mean. No MRI data exist. Injection site reactions were reported as 2.4 %. There was 27 % dropout. Class II Category C.

IVIG-- insufficient quality to be evaluated

Mitoxantrone randomized MRI blinded patients with aggressive MS EDSS <> 1.0. Entry 2 relapses in 2 years. EDSS outcome was better in treatment group (7 % progresses v. 37 % changed by one point over two years; however there was no change in MEAN change in EDSS score. Relapse outcome over 2 years was .89 v. 2.62. Percent relapse free was 63 v. 21 % all significant. NTT was 3.3. Class II/III Category C study due to trouble blinding.

MIMS MTX in SPMS bseline EDSS 4.5-7 with one point deterioration in year prior to entry. Primary endpoint was a composite of EDSS, Ambulation Index, number of relapses requiring steroids, and Standard Neurologic Status. The result was highly significant p<0.0001). There was a difference in groups on EDSS progression, at 6 months, mean number of treated relapses, time to first treated relapse, percentage of patients without a relapse, totoal number of relapses, and number of hospitalizations. Class II Category B. Key point is that it was ana actively relapsing group in the two years before study entry.

Secondary progressive MS:
Interferon B1b (European) entry 3-6.5 EDSS 70 % of patients had relapses in 2 years before study entry. Progression of disability was 49.7 % v. 38.9 % p=0.0048, also less patients became wheelchair bound. Relapses were reduced by about 30 %. No treatment effects in patients with baseline EDSS 6.0 or 6.5. Decreased T2 burden. NTT was 9.2 patients for 3 years to prevent one point EDSS increase. Class I Category A. Difference with SPECTRIMS (North American study) might be due to less pre treatment relapses in N American group.

Head to Head
INCOMIN Beta 1b v. Beta 1a (Betaseron v. Avonex) 2 year trial, primary endpoint was no of relapse free patients, and number of patients with no new T2 lesions on their MRI. Entry 2 relapses in 2 years, EDSS 1-3.5. EDSS progression of one point was 13 % over 2 years with Betaseron, 30 % with Avonex. T2 lesion burden showed o new lesions in 55 %with Betaseron, 26 % of those with placebo. NTT to make a patient relapse free for 2 years was 7 for betaseron v. avonex. Class I for MRI, Class III for clinical outcomes (unblinded)


EVIDENCE Rebif v. Avonex. primary endpoint was proportion of patients who were relapse free at 24 weeks. entry edss 0-5.5, pastients who got Rebif received 44 mcg tiw dose. Evaluating neurologists were blinded. There was no change in disability. Relapse free patients favored Rebif (74.9 v. 63.3 %, odds ratio 1.9, p=0.0005). Combined unique active lesions on MRI favored Rebif. NTT was 9 at 24 weeks and 10 at 48 weeks. Class I study.

Neutralizing Antibodies against Natalizumab

editorial Freedman MS, Pachner AR. Neutralizing antibodies to biological therapies: A "touch of grey" vs. a "black and white" story. Neurology 2007; 69:1386-1388.
based on article Calabresi PA, Giovannoni G, Confavreux C et al. The incidence and significance of antinatilizumab antibodies : results from AFFIRM and SENTINEL. Neurology 2007; 69: 1391-1403.

Cut to the chase:
from the editorial "With natalizumab, NAb's develop quickly, are readily measured, and can be shown to correspond with a loss of activity against a validated disease measure (eg. MR). With IFN B, NAbs develp more slowly, can spontaneously resolve and reappear depedning on titer and are not completely time synchronous with their effect on disease measures. " NAbs correlate (with natalizumab) with new MRI lesions, increased number of relapses, and progression of disability. Editorial notes the gray areas: some patients have NAbs and no re-emergence of disease, relapse rates are only higher inthe persistent NAbs group, and the persistent groups has more infusion reactions early, in the first few months of therapy, before its possible to identify the persistent NAbs group. The lack of persistence in AB might be due to antibodies to the antibodies. NAbs titers are not important.

In the Calabresi article, authors note that 9 % of patients in AFFIRM had NAbs, 3 % transiently and 6 % persistently. SENTINEL results were similar to AFFIRM.