Article Text

Download PDFPDF

Research paper
Autologous haematopoietic stem cell transplantation: a viable treatment option for CIDP
  1. R Press1,
  2. H Askmark2,
  3. A Svenningsson3,
  4. O Andersen4,
  5. H W Axelson5,
  6. U Strömberg6,
  7. A Wahlin7,
  8. C Isaksson7,
  9. J-E J Johansson8,
  10. H Hägglund9
  1. 1Department of Neurology, Clinical Neuroscience, Karolinska Institutet, Karolinska University Hospital Huddinge, Stockholm, Huddinge, Sweden
  2. 2Department of Neuroscience, Neurology, Uppsala University Hospital, Uppsala, Sweden
  3. 3Department of Pharmacology and Clinical Neuroscience, Umeå University, Umeå, Sweden
  4. 4Institute of Neuroscience and Physiology, University of Gothenburg, Gothenburg, Sweden
  5. 5Department of Neuroscience, Clinical Neurophysiology, Uppsala University Hospital, Uppsala, Sweden
  6. 6Department of Medical Sciences, Uppsala University and Section of Hematology, Uppsala University Hospital, Uppsala, Sweden
  7. 7Department of Radiosciences, Umeå University, Umeå, Sweden
  8. 8Department of Hematology and Coagulation, Sahlgrenska University Hospital, Gothenburg, Sweden
  9. 9Department of Hematology, Karolinska Institutet, Karolinska University Hospital, Stockholm, Huddinge, Sweden
  1. Correspondence to Dr Rayomand Press, Department of Neurology, Karolinska University Hospital Huddinge, Stockholm, Huddinge SE-14186, Sweden; rayomand.press{at}ki.se

Abstract

Objective Only 70–80% of patients with chronic inflammatory demyelinating polyneuropathy (CIDP) respond satisfactorily to the established first-line immunomodulatory treatments. Autologous haematopoietic stem cell transplantation (AHSCT) has been performed as a last treatment resort in a few therapy-refractory cases with CIDP. We describe the results of AHSCT in 11 consecutive Swedish patients with therapy-refractory CIDP with a median follow-up time of 28 months.

Method Case data were gathered retrospectively for AHSCT treatments in 11 patients with CIDP refractory to the first-line immunomodulatory treatments, intravenous high-dose immunoglobulin, corticosteroids and plasma exchange and to one or more second-line treatments used in 10 of the 11 patients.

Results The median Inflammatory Neuropathy Cause and Treatment (INCAT) score within 1 month prior to AHSCT was 6 and the Rankin score 4. Total INCAT and Rankin scores improved significantly within 2–6 months after AHSCT and continued to do so at last follow-up. The motor action potential amplitudes (CMAP) improved already within 4 months (median) after AHSCT. Three of the 11 patients relapsed during the follow-up period, requiring retransplantation with AHSCT in one. Eight of the 11 patients maintained drug-free remission upon last follow-up. AHSCT was safe but on the short term associated with a risk of cytomegalovirus (CMV) and Epstein–Barr virus reactivation, CMV disease, haemorrhagic cystitis and pancreatitis.

Conclusions Our results though hampered by the limited number of patients and the lack of a control group suggest AHSCT to be efficacious in therapy-refractory CIDP, with a manageable complication profile. Confirmation of these results is necessary through randomised controlled trials.

  • NeuropathY
  • Neuromuscular
  • Neuroimmunology
  • Neurophysiology

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Chronic inflammatory demyelinating polyneuropathy (CIDP) is a chronic progressive or relapsing-remitting sensorimotor neuropathy which is a consequence of autoreactive immunity targeting myelin in the peripheral nervous system (PNS).1 First-line treatments in CIDP consist of intravenous high-dose immunoglobulin (IVIg), corticosteroids and plasma exchange (PE).2 ,3 Approximately 70% of patients with CIDP respond adequately to one of the first-line treatments, and a further 10% upon switching of therapies.4 Of those who respond to one of the first-line treatments, 12–29% eventually become treatment-refractory.5 ,6 If not treated efficiently, secondary axonal loss progresses over time in the PNS, increasing the level of morbidity and risk of mortality.7 Severe disability despite treatment is seen in 13% of patients with CIDP.7 ,8 The mortality rates of CIDP, which vary between 4% and 17%, are due to respiratory failure or pulmonary embolism mainly in elderly patients.5 ,8 Patients with treatment-refractory CIDP usually receive second-line immunomodulatory drugs such as cyclophosphamide, ciclosporin, rituximab, azathioprine and methotrexate, although evidence for the efficacy of these treatment choices is lacking.2 Autologous haematopoietic stem cell transplantation (AHSCT), which is increasingly used for treatment of autoimmune disorders refractory to conventional immunosuppression, is an emerging treatment option in the field of therapy-refractory CIDP. AHSCT induces marked immunosuppression, presumably by reducing the number of autoreactive lymphocytes, expanding the pool of naive T cells and altering T cell diversity.9

Complete remission has been claimed with AHSCT in myasthenia gravis, systemic sclerosis and multiple sclerosis (MS).10 ,11 The use of AHSCT in therapy-refractory CIDP was first reported in 1992. The response was maintained until 5 years post-transplant when symptoms recurred.12 ,13 The results of AHSCT in treatment-refractory CIDP have been generally favourable even in more recent reports. Several subsequent reports of AHSCT treatment have shown long-term remission of the CIDP.14–18 Taken together, we have identified eight published cases of AHSCT used in treatment-refractory CIDP at seven different global centres.

We hereby report on clinical and neurophysiological results of 11 consecutive patients with therapy-refractory CIDP treated successfully with AHSCT in Sweden. To our knowledge, this is the largest patient series undergoing AHSCT for CIDP ever reported in a consequent manner.

Materials

The 11 therapy-refractory patients with CIDP in this study received AHSCT in the period September 2002–September 2012 at four different University Hospitals in Sweden. Patients with CIDP with a high level of functional impairment as determined by the treating clinician and not improving functionally or suffering from multiple relapses despite receiving immunomodulatory treatments were deemed therapy-refractory.

Clinical characteristics of the patients are presented in table 1. Ten of the 11 patients were men. One of the 11 patients (no. 1) has previously been reported in a publication.16

Table 1

Demographics, clinical and electrophysiological data as well as functional scores in therapy-refractory patients with CIDP prior to treatment with AHSCT

All patients had definite typical CIDP according to the clinical and electrophysiological European Federation of Neurological Sciences (EFNS) 2006 criteria.19 Each of the 11 patients fulfilled a minimum of two of the seven electrophysiological EFNS criteria for CIDP (table 1). The electrophysiological data were re-evaluated retrospectively for the six patients who had been diagnosed prior to the publication of the EFNS 2006 electrophysiological criteria and were found to fulfil these criteria.

All patients had motor symptoms in the lower as well as upper extremities. Proximal leg muscles were involved in all patients. None of the patients had pure motor or sensory CIDP, pure distal symptoms or significant asymmetrical features. Only one of the 11 patients (no. 2) had a monoclonal gammopathy (IgM-κ), but no myelin-associated glycolipid (MAG) serum antibodies. This patient had no signs of lymphoproliferative disease upon haematological investigation. The fact that this patient had a motor-dominant neuropathy, no anti-MAG serum antibodies and electrophysiological signs of demyelination not only restricted to distal segments of the peripheral nerves indicated CIDP and not a paraproteinemic neuropathy in this case. None of the patients had clinical signs of polyneuropathy organomegalia endocrinopathy M-protein and skin changes syndrome.

One patient (no. 6) had MS in addition to CIDP, with the onset of first MS symptoms preceding those of CIDP by 3 years. This patient fulfilled both the clinical and radiological criteria for MS, as well as clinical and electrophysiological criteria for CIDP (see case description below). One patient (no. 7) had been judged to have relapsing Guillain–Barré syndrome with a total of four episodes of acute polyradiculoneuritis during the 13-year period preceding onset of CIDP.

All of the 11 patients had been judged refractory to first-line treatment options (only nine were treated with PE), and 10 of the 11 patients had also been treated with one or more second-line immunomodulatory agents but without response (table 2). The reason why a single patient (patient no. 2) was not treated with any second-line treatment options prior to AHSCT was the severity of the disease and its fast rate of progression.

Table 2

Immunotherapy prior to AHSCT as well as follow-up and long-term outcome after AHSCT in therapy-refractory patients with CIDP

Methods

This study was based on retrospective case data collection on all AHSCT treatments for patients with CIDP performed to date in Sweden.

Clinical and neurophysiological evaluation

Clinical data were gathered retrospectively for all patients except patient no. 11. Disability prior to and following AHSCT was measured using the 0–10 point INCAT scale3 ,20 as well as the 0–6 point modified Rankin (mRANKIN) scale.21 In patients examined prior to the introduction of the INCAT score in 2001, the INCAT scores were recreated retrospectively using chart data obtained from treating neurologists, physiotherapists and occupational therapists (table 1).

The median, ulnar, peroneal and tibial nerves were stimulated proximally (at elbow or knee level) and compound muscle action potentials (CMAPs) were recorded from hand or feet muscles. A mean compound muscle action potential (mCMAP) amplitude was calculated for each subject by taking the average of CMAP amplitudes for all stimulated nerves22 before and after AHSCT. Needle electromyography EMG was performed in the lower and upper extremities in all patients. Upper and lower extremity nerves were investigated bilaterally in seven subjects and unilaterally in four.

Haematopoietic stem cell transplantation procedures

Peripheral blood stem cells (PBSCs) were mobilised by high-dose cyclophosphamide (Cy) 2–4 g/m2 (n=9) on day zero, followed by granulocyte colony stimulating factor (G-CSF) 10 μg/kg (n=10) from day four to end of apheresis, and in addition plerixafor 0.2 mg/kg in one patient from day 11 to 12. Two patients were mobilised with rituximab 375 mg/m2 on the day before starting G-CSF, which was then given at a dose of 16 μg/kg for 4 days. In the retransplanted patients, PBSCs were mobilised with Cy 2 g/m2 and G-CSF 10 μg/kg from day 4 to 11. Apheresis was initiated at day 11 if the number of CD34 cells in peripheral blood was >20×106 (table 3). The aim was to harvest >4 × 106 CD34 cells.

Table 3

Treatment protocols and procedure-related complications in patients with therapy-refractory CIDP treated with AHSCT

Conditioning regimens and transplantation

The conditioning regimen consisted of Cy 35–50 mg/kg alone (n=7), melphalan 200 mg/m2 (n=1) or carmustine, cytarabine, melphalan and etoposide (BEAM) in three patients. Cyclophosphamide alone was used in the retransplanted patient. Antithymocyte globulin (ATG) was used in nine patients: thymoglobulin for 2 days (total dose 5–10 mg/kg) in six of the nine patients and ATG-Fresenius for 3 days (total dose 30 mg) in the other three. In the retransplanted patient, ATG-Fresenius was given. All grafts were unmanipulated, that is, not CD34-selected (table 3).

Statistics

All statistics were performed with non-parametric analysis methods. Comparison of disability scores over the course of CIDP was performed with one-way Analysis of variance (ANOVA) and the non-parametric Kruskal–Wallis multiple comparison test. mCMAP levels before and after haematopoietic stem cell transplantation (HSCT) were compared with the Wilcoxon matched-pairs signed rank test. p Values were significant if <0.05.

Ethical permission

This article is based on retrospective data about patients with CIDP undergoing AHSCT treatment during a 10-year time period prior to this study. The decision to provide AHSCT on an experimental basis for the therapy-refractory condition was based purely on clinical grounds in each individual case, and not influenced by this study. Hence, no ethical permission or informed consent was deemed necessary at the time of AHSCT.

Results

Clinical data

The mean age at time of AHSCT treatment was 55 years (range 23–68). The median duration of disease and of treatment period prior to the AHSCT is shown in table 4. Long-term follow-up data (≥18 months post-AHSCT) were available for 10 of the 11 patients, whereas one patient was only followed up for 6 months. The median duration of follow-up after AHSCT was 28 months (6–127 months) (table 4).

Table 4

Raw data for clinical scores and CMAP values prior to and following AHSCT in 11 therapy-refractory patients with CIDP

The patients had a relatively high level of functional impairment prior to AHSCT as reflected by the total INCAT median score of 6 and median mRANKIN score of 4 within 1 month prior to the 12 AHSCT occasions (patient no. 1 received AHSCT twice due to relapse) (table 4). Disability improved quite rapidly during the 2–6-month time period following AHSCT. At last follow-up, the INCAT and mRANKIN levels continued to be lower than pretreatment values (table 4 and figure 1A,B).

Figure 1

Functional scores and mean compound muscle action potentials in patients with chronic inflammatory demyelinating polyneuropathy (CIDP) prior to and after autologous haematopoietic stem cell transplantation (AHSCT). Total INCAT (A) and modified Rankin scores (B) measured over the course of disease in 11 patients with therapy-refractory CIDP. Patients were examined within 4.5 months (median) prior to the 12 AHSCT occasions (=Pre-Treatm), within 4 months after AHSCT (Post-Treatm 1) and at last follow-up 28 months (median) after AHSCT (Post-Treatm 2).

Immunomodulatory treatments were discontinued in four of the 11 patients immediately following AHSCT and in an additional four within 1 year after treatment. During the long-term follow-up, only three of the 11 patients relapsed. The relapses occurred at 23 months (patient no. 1) and 14 months (patient nos. 5 and 10) after AHSCT (table 3). The relapses were successfully treated with a new course of AHSCT (patient no. 1), tocilizumab (patient no. 5) and low-dose prednisolone (patient no. 10). The reason why AHSCT was not used for treating the relapses in patient nos. 5 and 10 was lack of patient acceptance for a renewed ASHCT in patient no. 5 and the swift improvement of symptoms with oral steroid therapy in patient no. 10. At last follow-up, eight of the 11 patients were in remission off-treatment, whereas three of the 11 patients needed treatment (tocilizumab in pat. 5 and oral steroids in pats. 9 and 10) to maintain a stable status.

Neurophysiological data

Neurophysiological data were available for all the 11 patients prior to, as well as after AHSCT, for a total of 12 AHSCT occasions, since one patient had received AHSCT on two occasions due to relapse. The evolution of mCMAP and simultaneously measured INCAT levels over the course of CIDP are shown for each individual patient in figure 2. Post-AHSCT mCMAP levels (median 1.88 mV) were higher than pretreatment levels (median 0.84 mV) (table 4).

Figure 2

Evolution of INCAT and mean compound muscle action potential (mCMAP) scores over the course of chronic inflammatory demyelinating polyneuropathy (CIDP). The mCMAP measured simultaneously with total INCAT scores over the course of CIDP in the 11 patients prior to autologous haematopoietic stem cell transplantation (AHSCT) as well as afterwards. mCMAP levels were calculated using the compound muscle action potential values of the ulnar, median, tibial and peroneal nerves obtained by proximal stimulation and distal registration of each nerve. * ,  time point of AHSCT.

Safety and medical complications of AHSCT

Early complications were seen in six of 12 transplantation occasions (table 4). Patient no. 1 had cytomegalovirus (CMV) and Epstein–Barr virus (EBV) reactivation after retransplantation, which resolved spontaneously. Patient no. 4 developed an Escherichia coli bacteraemia and BK-virus-positive haemorrhagic cystitis. Patient no. 7 suffered from coagulase-negative Staphylococci bacteraemia, BK virus-positive haemorrhagic cystitis and CMV reactivation. In patient no. 8, Klebsiella, Pseudomonas and α-Streptococci septicaemias were diagnosed requiring treatment in the intensive care unit. This patient also developed a transient CMV colitis. Patient no. 10 had neutropenic fever and patient no. 11 pancreatitis as well as CMV reactivation. Long-term, but transient consequences were diagnosed in four cases, hypothyreosis in patient no. 4, long-lasting neutropenia that normalised after 9 months in patient no. 5, relapsing Clostridium difficile-positive diarrhoea in patient no. 8 and anaemia of unknown origin that recovered spontaneously in patient no. 10 (table 4).

Illustrative cases

Patient no. 1 Relapsed after first AHSCT but was successfully retreated with AHSCT.

This patient presented with a motor-dominant CIDP, characterised by progressive tetraparesis and areflexia, but no subjective sensory symptoms in August 2001 (table 1). Treatment with IVIg every 3 weeks resulted in some improvement of ambulation, but in 2002 the patient deteriorated rapidly despite adding on steroids, azathioprin and PE and was bedridden. AHSCT was performed in September 2002. The patient subsequently improved and 4 weeks after the last PE, gait and muscle strength were normal. Treatment-free remission was maintained until October 2004 when he suffered a partial relapse, necessitating retransplantation with AHSCT. The remission was maintained even at last follow-up in March 2013.

Patient no. 4 responding only temporarily to alemtuzumab, but achieved long-term remission after AHSCT.

This patient presented in 2005 with predominantly distal sensorimotor CIDP in all extremities as well as arm tremor. IVIg, and later on addition of oral prednisolone plus methylprednisolone (IVMp) resulted in significant improvement. IVIg treatment was subsequently replaced by oral cyclophosphamide because of thromboembolic complications. Due to continued fluctuating symptoms, the patient received a course of high-dose cyclophosphamide (2 g/m2), followed later on by 60 mg alemtuzumab, resulting in a transient improvement. Treatment with AHSCT was decided on, but was delayed until December 2009 due to alemtuzumab-induced bone marrow suppression. The neurological symptoms improved rapidly 3 months after AHSCT; hence, the patient could taper all steroids. The patient maintained a treatment-free remission upon last follow-up in May 2012.

Patient no. 6 Double diagnosis of MS and CIDP, developed CIDP while on natalizumab treatment, with both conditions responding to AHSCT.

This patient had an episode of unilateral optic neuritis in 2001, followed later by relapses whereby the clinical and radiological criteria for MS were fulfilled. In the beginning of 2003, the patient developed a progressive weakness and sensory symptoms in all extremities and areflexia in the legs. Cerebrospinal fluid (CSF) analysis and neurophysiological evaluation confirmed the diagnosis of CIDP according to the Saperstein criteria.23 She was hence treated with monthly pulsed IVMp followed by IVIg until 2005 when the treatment was switched to high-dose interferon β. Due to a relapse both of MS and CIDP despite IVMp, she was in 2007 switched to monotherapy with natalizumab. In mid-2009, the patient deteriorated with regard to CIDP, which could not be reversed by the addition of IVIg treatment. AHSCT was finally performed in April 2010. In June 2010, she was able to walk independently and continued to improve after that until no apparent residual symptoms remained by the last examination November 2012.

Patient no. 9 Positive but only partial response to AHSCT probably due to extensive axonal damage (denervation activity on EMG prior to AHSCT).

This patient was diagnosed with CIDP due to a 2.5-month history of progressive tottering gait, distal numbness in all extremities, a unilateral peripheral facial palsy and areflexia in January 2011. Regular treatment with IVIg 2 g/kg, and later on with IVMp as well as oral steroids resulted only in a transient response, and the patient became a wheelchair user by June 2011. Treatment with rituximab at a weekly dose of 750 mg/m2/week for 4 weeks did not lead to improvement. EMG at this time showed signs of denervation activity in proximal and distal muscles of all extremities. AHSCT was performed in September 2011. On examination 7 weeks later the patient could walk without support, despite bilateral foot drops. At follow-up in March 2013 when the patient was still on prednisone 5 mg/day, a further improvement of muscle strength was noted though there was still pronounced weakness of the finger movements and the feet.

Discussion

We present the largest yet documented series of patients with therapy-refractory CIDP examined over the disease course, prior to and following AHSCT, including a relatively long follow-up period of 28 months. Data were in most cases gathered retrospectively.

All patients included were deemed therapy-refractory to established first-line treatments for CIDP and in addition, 10 of the 11 patients who were treated with a number of second-line treatments continued to be treatment-refractory prior to AHSCT. Using INCAT and mRANKIN scores, we measured a significant functional improvement already 2–6 months after AHSCT treatment. The improvement was observed even in severely disabled patients who in most cases continued to improve when seen at last follow-up, suggesting a long-lasting remission. The clinical improvement was paralleled by an improvement of CMAP values 4 months (median) after AHSCT. Interestingly the post-AHSCT clinical improvement as measured by INCAT and mRANKIN scores was seen even in patients with the lowest pre-AHSCT mCMAPs, which may be due to the fact that low CMAPs are more readily caused by proximal conduction failure rather than axonal loss.

During the follow-up, only three of the 11 patients had relapsed and when last seen, the majority (eight of the 11) of patients were in a treatment-free remission. The relapses were treated successfully with a new course of AHSCT, tocilizumab and low-dose prednisolone, respectively. Two of the three patients who relapsed had been treated with a cyclophosphamide conditioning (in one case together with ATG), but the third patient had been treated with a BEAM conditioning. We did not succeed in identifying any specific clinical or HSCT protocol-related parameter that sets the relapsing patients apart from those who did not relapse following AHSCT.

Our results, although restricted to only a few cases, also indicate that the monoclonal antibodies alemtuzumab and natalizumab may not be effective in treatment of advanced CIDP. One patient relapsed short time after receiving alemtuzumab treatment and a second patient experienced a severe progression of CIDP while actually being treated with natalizumab for MS. Our observation with regard to the lack of efficacy of natalizumab in CIDP supports a previous report by Wolf et al.23

In general, AHSCT was relatively well tolerated, but there was some morbidity both early and late after transplantation. CMV reactivation, but not CMV disease following AHSCT for multiple myeloma is relatively common, and approximately half of the studied patients developed CMV reactivation.24 Only one of the 12 transplantation occasions led to a CMV colitis. EBV reactivation is uncommon after AHSCT for malignant diseases. The addition of ATG to high-dose chemotherapy may impose increased risk of EBV reactivation due to immunosuppression.25 In our series, one transplantation occasion led to reactivation of EBV, but the patient did not develop EBV post-transplantation lymphoma. Pancreatitis is usually not seen after AHSCT, but was diagnosed in one patient. Long-term neutropenia and anaemia in two patients were probably caused by long-term chemotherapy before AHSCT. Our data cannot determine which of the two AHSCT protocols, cyclophosphamide or BEAM, is to be preferred from a risk–benefit ratio standpoint. Our results indicate that the efficacy of both treatment protocols seems to be similar.

Our study is limited by the lack of a control arm, relatively low number of patients as well as its retrospective nature that did not permit uniform intercentre follow-up procedures. Also, INCAT scores were in some cases re-created from chart data, which may raise an issue of accuracy of estimates. However, this was in part compensated by the available data on the level of impairment in the upper as well as lower extremities as carefully documented by occupational therapists and physiotherapists in those cases that lacked INCAT and mRANKIN scores in patient charts.

Considering the relatively good tolerability of AHSCT and its capability to achieve long-term remission of CIDP, this treatment method may be an option at an earlier stage than the traditional ‘last treatment resort’, when a significant secondary axonal loss has already occurred. We suggest that AHSCT may be considered in patients with CIDP already at a time point when the patient is judged to be therapy-unresponsive to first-line treatments, defined as lack of improvement by ≥1 point in total INCAT scores after 1–2 courses of IVIg and 2–3 months of corticosteroid treatment±PE, or inability to reduce the total INCAT score to ≤4 after 6–12 months of immunomodulatory treatment.

Although suffering from the above-mentioned limitations, the results of this study suggest AHSCT may be a treatment that offers long-term remission even in severe treatment-refractory CIDP with an overall acceptable side effect and risk profile. However, the efficacy of AHSCT in patients with CIDP remains to be confirmed by randomised controlled trials.

References

Footnotes

  • Contributors RP is mainly responsible for gathering clinical data from the treatment centres, and neurophysiological data from HWA, composing all data, performing the statistics, creation of tables and figures (except for table 3) and drafting of the manuscript, except for the methods and results sections dealing with AHSCT protocols and complication, as well as the case reports (named ‘illustrative cases’ in this manuscript) of individual patients from the other three centres in Sweden. He was also the physician in charge of the two patients who underwent AHSCT at Karolinska University Hospital, and personally responsible for the follow-up of those patients. HA is the physician in charge of the five patients who underwent AHSCT at Uppsala University Hospital, and personally responsible for the follow-up of those patients. He provided all the clinical data for the patients included from Uppsala and wrote the case reports for the illustrative cases from Uppsala. He took a pivotal part in the design of the manuscript as well as scrutinising the many versions of the manuscript before its completion. AS is the physician in charge of the three patients who underwent AHSCT at Umeå University Hospital, and personally responsible for the follow-up of those patients. He provided all the clinical data for the patients included from Umeå and wrote the case reports for the illustrative cases for those three patients, not all of which are included in the final manuscript. AS also assisted in an extensive review of the manuscript and its figures prior to it´s final submission. OA is the physician in charge of the patient who underwent AHSCT at Gothenburg University Hospital, and personally responsible for the follow-up of that patient. He provided all the clinical data for his patient and wrote the case report for the illustrative case from Gothenburg. He took an active part in the design of the manuscript as well as scrutinising the many versions of the manuscript before its completion. HWA was the responsible specialist in neurophysiology who analysed all neurophysiological data gathered from the study centres, and calculated mCMAP values for all the 11 included patients. He contributed to the original version of the figures comparing INCAT values with mCMAP (figure 1). He contributed to the design of the manuscript and took an active part in scrutinising the many versions of the manuscript before its completion. US is the haematologist responsible for the AHSCT treatment of the five patients from Uppsala. She provided structured data regarding the treatment protocols and treatment-related complications for the patients in Uppsala. She took an active part in the design of the manuscript as well as scrutinising the many versions of the manuscript before its completion. Both AW and CI work at the Department of Haematology at Umeå University Hospital. They were jointly responsible for the AHSCT treatment of the three patients from Umeå. They provided equally structured data regarding the treatment protocols and treatment-related complications for the patients in Umeå. Both AW and CI took an active part in the design of the manuscript as well as scrutinising the many versions of the manuscript before its completion. J-EJJ is the haematologist responsible for the AHSCT treatment of the single patient from Gothenburg University Hospital. He provided structured data regarding the treatment protocol and treatment-related complications for this patient. He took an active part in the design of the manuscript as well as scrutinising the many versions of the manuscript before its completion. HH was one of the colleagues who initiated the idea of structurally reporting the results of AHSCT treatment of patients with CIDP in Sweden to the scientific community. He is also the haematologist responsible for the AHSCT treatment of the two patients from Karolinska University Hospital, and hence provided data regarding the treatment protocols and treatment-related complications for the two patients. He was also the coordinator for the haematological data, that is, was responsible for gathering of all AHSCT treatment and complications data from his fellow ha ematologists and compiling the data in table format (table 3), as well as in the body of the manuscript. Besides writing part of the manuscript HH took an active part in its overall design as well as scrutinising the many versions of the manuscript before its completion.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement There do not exist any remaining data from this study, which are not reported in this manuscript.