Autologous Hematopoietic Stem Cell Transplantation in Autoimmune Diseases

Claudio Annaloro; Francesco Onida; Giorgio Lambertenghi Deliliers


Expert Rev Hematol. 2009;2(6):699-715. 

In This Article

Systemic Sclerosis

Systemic sclerosis is a generalized AD characterized by diffuse, disabling skin thickening combined with fibrotic changes in multiple organs, notably heart and lung, leading to their end-stage failure. Despite considerable evidence of autoimmunity, the sequence of events leading to fibrosis still awaits to be entirely elucidated. A peculiar feature of SS is the lack of a definite treatment strategy able to block disease progression and, even less, to reverse fibrotic changes. Although it has been claimed that some benefit can be expected by the use of pulsed cyclophosphamide,[71] the matter is debated, at best, and life expectancy is shortened, with a median 5-year survival that may not exceed 50% in rapidly progressive forms.[72]

These disappointing results have increased the interest in HSCT as a possible way of modifying the natural history of the disease. Among rheumatologic ADs, SS is the most common indication, but there is some doubt as to whether the interest in autologous HSCT has been elicited by distinctive features of the disease or by the lack of suitable therapeutic alternatives. An aid to the development of autografting strategies was the definition of the Rodnan skin score (RSS) as a means to define disease extension and response to therapy.[63] Patients with SS are generally referred for autologous HSCT because of extensive skin involvement or of a rapidly progressive course. At that time, most patients have extensive cardiopulmonary compromise, testified by impaired CO diffusion (DLCO), reduced left ventricular ejection fraction (LVEF) and pulmonary hypertension.

After various preliminary experiences, a homogeneous series of autografted SS patients was reported in 2002 by the Fred Hutchinson Cancer Research Center (FHCRC),[47] where 19 patients with poor-prognosis disease were accrued. After a TBI/CTX conditioning regimen combined with equine ATG, they received G-CSF mobilized, positively selected CD34+ cells. Patients were considered at poor prognosis if they had a 50% 5-year life expectancy, defined by high RSS and organ dysfunction (lung, heart and/or kidney) or by rapidly progressive clinical course. Results focused on the somewhat contradictory aspects of SS lung disease. Most patients with advanced skin disease had lung involvement as major organ dysfunction. On the other hand, lung toxicity was linked both to high TRM (three out of 19) and to post-HSCT disease progression. Finally, a durable improvement in RSS was observed in all evaluable patients, whereas no case of improvement in lung function could be observed. Although not recent, this study encompasses most of the debated issues on the field of autologous HSCT in SS. Similar results were reported in a multicenter French study where immunoablative chemotherapy was used instead of TBI.[73]

Subsequent experiences were based on comparable-size single institution experiences[46] and on multicentric retrospective surveys, notably by the EBMT,[74] both types of reports having obvious limitations. In spite of the lack of comparative studies, some development in HSCT strategies can be outlined. CTX (2–4 g/m2) has been regularly added to G-CSF in the mobilizing schedule and positive CD34 selection has always been chosen as the favorite approach of harvest manipulation. Since this procedure leads anyway to a suboptimal lymphocyte depletion, ATG was commonly held as unavoidable to achieve the desired pretransplant reduction in mature lymphocytes.

Patients selection criteria have been perceived as a key step to improve results in SS, since TRM exceeded 10% in the first reports, a value significantly above the corresponding one in oncohematological patients.[75,76] Apart from different delineations of the required RSS, much effort has been devoted to better define organ dysfunction. Patients with a LVEF below a mean threshold value of 45% should be discarded. Lung function should be evaluated by combining spirometric and DLCO values, with the latter parameter below 45% being an exclusion criterion. An additional, widely accepted cause of exclusion is a pulmonary artery pressure (PAP) exceeding 50 mmHg.[77] The application of proper selection criteria has probably played a primary role to reduce TRM to the levels of neoplastic HSCT recipients in the more recent reports.[9,78]

Another major issue concerns the choice of the conditioning regimen. The FHCRC experience underscored the relevance of lung shielding in delivering TBI in order to reduce TRM.[47] Subsequently, less toxic regimens have been progressively substituted for TBI in order to reduce lung toxicity and CTX 200 mg/kg has gained the best acceptance.[79]

Concerning results, an improvement in RSS was a common finding and post-HSCT progression of skin disease was rare.[57,80] Conversely, it has been widely demonstrated that no improvement in organ dysfunction can be achieved; even recently, Vonk et al. defined as "not very successful" the results of autologous HSCT on lung disease.[46] Optimization of selection criteria and conditioning regimens may only lead to a further reduction in TRM and in iatrogenic lung damage.[57] This may translate into some improvement in QoL but casts some doubt about the effective gain in survival. Therefore, some additional perplexity arises as to whether autologous HSCT in SS is really a cost-effective therapeutic procedure.

The evolution of therapeutic concepts are summarized in the ongoing Phase III EBMT Autologous Stem cell Transplantation International Scleroderma (ASTIS) trial, comparing autologous HSCT with monthly pulse CTX as a reference treatment. Selection criteria mirror the aforementioned discussion and exclude patients with low LVEF/DLCO and high PAP. Conditioning regimen choice reflects the progressive substitution of high-dose CTX for TBI. The choice of this reference treatment probably reflects the complexity to find a suitable one. Similar selection criteria and conditioning regimen were used in the report by Vonk,[46] one of the most recent homogeneous experiences. Although a precise comparison with the report by McSweeney[47] is not possible, in the Vonk's study[46] TRM fell from three out of 19 to one out of 26; overall survival chance rose from approximately 80% at 3 years to 96.2% at 5 years; after a longer follow-up, disease progression occurred in four out of 24 evaluable patients in comparison with three out of 16 reported by McSweeney, with almost all evaluable patients achieving some degree of skin response in both series. Conversely, no benefit in organ involvement was documented in both experiences, since autologous HSCT is commonly believed to induce stabilization of visceral involvement in SS.

A 6-year evolution of HSCT in ADs, has apparently led to an improvement in TRM and survival. Nevertheless, the question may be raised as to whether the better results simply mirror evolving selection criteria: patients treated in 2002 had a life expectancy not exceeding 50% at 5 years, whereas patients treated in 2008 had a more favorable prognosis. Moreover, the improvement in TRM rate after autologous HSCT for ADs and notably for SS should not be overemphasized also considering that most patients are female, whereas oncohematological series encompass a majority of male patients. Although TRM after HSCT is a rather rare event and studies on this issue are rather old, female gender has been found on large series to be a favorable prognostic factor.[81] Possibly, the observed TRMs in ADs should be corrected for gender. Possible specific complications of autologous HSCT, as secondary leukemia, have not been reported. Anyway, a significant progress over the last few years is hard to recognized. Moreover, no method has been identified to improve organ dysfunction, with lung progression being the leading cause of death in patients with SS, although long-term disease stabilization cannot be disregarded in the absence of reliable alternative therapeutic options. Therefore, the question is still open whether autologous HSCT in SS represents a promising treatment strategy or a consequence of hopelessness. An argument favoring the former interpretation derives from experimental studies showing that autologus HSCT in SS patients bears the potential of inducing regression of the capillary loss, characteristically observed in SS patients not undergoing this procedure.[82] Along with this view, Miniati et al. observed a remodeling of capillary tree in SS patients undergoing autologous HSCT but not in those treated with pulsed cyclophosphamide.[71] Unfortunately, these findings do not translate as far in an improvement of visceral compromise.

On the opposite side, perplexity is indicated by the recent proposal of cell therapy modalities for SS alternative to autologous HSCT, such as the use of mesenchymal stem cells, having also an anti-inflammatory and tissue regenerative potential.[83,84] Although mesenchymal stem cell research has been gaining growing interest, the proposal seem to derive mainly from the "not very successful"[46] results of autologous HSCT. Apart from theoretical considerations and some preliminary laboratory evidence,[85] this way is still to be explored completely.[84]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.