Lights and Shadows of Cyclophosphamide in the Treatment of Multiple Sclerosis

Cyclophosphamide (cy) is an alkylating agent used to treat malignancies and immune-mediated inflammatory nonmalignant processes. It has been used as a treatment in cases of worsening multiple sclerosis (MS). Cy is currently used for patients whose disease is not controlled by beta-interferon or glatiramer acetate as well as those with rapidly worsening MS. The most commonly used regimens involve outpatient IV pulse therapy given with or without corticosteroids every 4 to 8 weeks. Side effects include nausea, headache, alopecia, pain, male and women infertility, bladder toxicity, and risk of malignancy. Previous studies suggest that cy is effective in patients in the earlier stages of disease, where inflammation predominates over degenerative processes. Given that early inflammatory events appear to correlate with later disability, a major question is whether strong anti-inflammatory drugs, such as cy, will have an impact on later degenerative changes if given early in the disease to halt inflammation.


Introduction
Multiple sclerosis (MS) is an acquired inflammatory immune-mediated disorder of the central nervous system, characterized by inflammation, demyelination, and primary or secondary axonal degeneration. It clinically manifests with signs of multiple neurological dysfunctions, followed by recovery or increasing disability. Cyclophosphamide (cy) (the generic name for Endoxan, Cytoxan, Neosar, Procytox, and Revimmune), also known as cytophosphane, is a nitrogen mustard alkylating agent from the oxazophorine group ( Figure 1). An alkylating agent adds an alkyl group (C n H 2n+1 ) to DNA. It attaches the alkyl group to the guanine base of DNA, at number 7 nitrogen atom of the imidazole ring. This leads to the synthesis of aberrant couples of cytosin-tymine. The DNA reparation system of the cells removes the modified guanine, triggering cell apoptosis. Cy is converted by mixed function oxidase enzymes in the liver to active metabolites. The main active metabolite is 4-hydroxycyclophosphamide, which exists in equilibrium with its tautomer, aldophosphamide. Most of the aldophosphamide is oxidised by the enzyme aldehyde dehydrogenase (ALDH) to make carboxyphosphamide. The intracellular level of this enzyme has been shown to be directly related to cellular resistance to activated cy and is believed to be important in the survival of cells capable of repopulating marrow in autologous bone marrow transplant procedures [1]. Both hematopoietic progenitors and intestinal crypt stem cells display high levels of cytosolic ALDH and are accordingly relatively resistant to cy. Tumor cell resistance to cy may also result from high cytosolic ALDH levels [2]. A small proportion of aldophosphamide is converted into phosphoramide mustard and acrolein. Acrolein is toxic to the bladder epithelium and can lead to hemorrhagic cystitis [3].
Cy and the related nitrogen mustard-derived alkylating agent ifosfamide were developed by Norbert Brock and ASTA (now Baxter Oncology). Brock and his team synthesised and screened more than 1,000 candidate oxazaphosphorine compounds. They converted the base nitrogen mustard into a nontoxic "transport form". This transport form was a prodrug, subsequently actively transported into the cancer cells. Once in the cells, the prodrug was enzymatically converted into the active, toxic form.
Cy is widely used, together with other antineoplastic drugs, for the treatment of leukemias, lymphomas, and carcinomas but also for the treatment of immune mediated diseases such as vasculitis (Wegener's granulomatosis, polyarteritis nodosa) [4,5], kidney diseases (lupus nephritis idiopathic nephrotic syndrome) [6], and for the treatment of severe systemic-onset juvenile rheumatoid arthritis [7] and interstitial lung disease associated with collagen vascular diseases [8]. Cy is also used for neurological diseases such as refractory cases of polymyositis or inflammatory neuropathies [9,10].
Cy was first tested in MS in 1966 [11]. Even if it is not a drug licensed for MS, due to the lack of adequate phase III studies, cy has been used for the treatment of selected MS patients who have had a partial response to previous treatment with the FDA-approved drugs. Furthermore, we think that cy could have been partially ignored by researchers and pharmaceutical companies due to the low cost of the drug. This could partly explain the lack of adequate randomized controlled studies. This paper will focus on the results obtained with cy to treat MS patients. In addition, it will report the different protocols of cy use, highlighting results and side effects.
MS is considered to be mediated by T helper type-1 (Th1) cells [21][22][23]. In humans, increased production of interferon-γ (IFN-γ) by peripheral blood mononuclear cells (PBMCs) has been shown to precede clinical attacks [24], and injection of recombinant IFN-γ induces exacerbations of the disease in patients with MS [25]. Cy acts on cellmediated and humoral immunity through its effects on both T and B cells. It has been shown to enter the nervous system, as it can be recovered from the cerebrospinal fluid of treated multiple sclerosis patients [26,27]. Among the FDA-approved drugs, currently used in MS, only fingolimod crosses the blood brain barrier and may, therefore, have direct effects on the CNS [28]; the relationship between cladribine (experimental drug) and the blood brain barrier is unclear, the concentration of cladribine in the cerebrospinal fluid is around 25% of what is available at the plasma level in patients without CNS disease [29]. Earlier studies in MS patients treated with cy had demonstrated a lymphopenia induced by cy involving both T and B cells [30,31] with a more pronounced effect on CD4 cells [32,33]. More recent studies showed that cy could also have selective effects on the immune system. Specifically, it increases Th2 cytokine such as IL-4, IL-5, IL-10, and TGF-β, a cytokine secreted by regulatory T cells (T-reg), and for this reason, cy is associated with eosinophilia [34]. Patients treated with cy showed an increased frequency of both MBP and PLP cells secreting IL-4 [35], while this effect was not observed in tetanus-toxoid-secreting cells and in MS patients treated with methylprednisolone [36]. The preferential shift towards a Th2-type pattern was also seen in terms of chemokine receptor expression. Chemokine receptors have been found to differentiate between polarized T helper type-1 (Th1) and type-2 (Th2) lymphocytes. The chemokine receptors CCR5 and CXCR3 are expressed primarily on Th1 cells and CCR3, CCR4, and CCR8 on Th2 cells. Previous studies of the expression of chemokine receptors in MS showed that active MS plaques are infiltrated by CCR5+ and CXCR3+ T cells that are major producers of IFN-γ. In MS patients, cy induced a marked increase in the percentage of CCR4+ T cells that produced high levels of IL-4 and reversed the increase in the percentages of IFN-γproducing CCR5+ and CXCR3+ CD8+ T cells [37].
IL-12 is a heterodimeric cytokine produced mostly by phagocytic cells and induces cytokine production, primarily IFN-γ, from T cells. Several studies in humans [38,39] and in mice [40,41] have assigned a role to IL-12 (linked to IL-23 and IL-17) as the promoter of Th1 cell generation, acting in antagonism with IL-4, the major promoter of Th2 responses. Administration of IL-12 to mice after the transfer of encephalitogenic cells resulted in increased severity and duration of EAE; treatment with anti-IL-12 antibodies substantially reduced the incidence and severity of adoptively transferred EAE [42]. Elevated serum levels of IL-12 as well as an increase in T cell receptor-mediated IL-12 secretion have been reported in the chronic progressive form of MS [43,44]. The results of these studies suggest that IL-12 could play an important role in the pathogenesis of EAE and MS.
In patients treated with methotrexate, methylprednisolone, or cy/methylprednisolone, only the last treatment normalized the elevated IL-12 production. Patients followed prospectively before and after starting CY/MP treatment, in fact, showed a gradual decrease in IL-12 and IFN-γ production and an increase in IL-4 and IL-5 [45,46].
In summary, probably a Th1-type cytokine bias has a role in the pathogenesis of the disease and is reversed by cy/MP treatment with an associated Th2 and TGF-β (Th3-Treg) type response.

Clinical Data
Several reports have been published on the clinical effects of cy in MS. Many reports showed that cy is effective in MS, but not all studies have shown positive effects. La Mantia et al. in 2007 conducted an evaluation of the published studies performed on cy in MS. Of the 326 identified references, 80 were selected for full review and only four RCTs were selected for the final analysis. The authors concluded that intensive immunosuppression with cy (alone or associated with ACTH or steroids) in patients with progressive MS compared to placebo or no treatment (152 participants) did not prevent the long-term (12-18-24 months) risk of evolution to a next step of the EDSS. However, the authors suggested that the lack of efficacy in progressive MS does not lead to the conclusion that the drug is not effective in the inflammatory phase of the disease [47]. Cy, in fact, is considered a treatment option in several recently published MS treatment guidelines [48]. However, a major question for physicians treating patients with MS is how cy should be used to obtain the best results and to avoid side effects. As we will see in the following paragraphs, the studies conducted since 1966 in MS patients give us some answers.

Early Studies
The first clinical trial with cy was published at the end of the 1950s [49].
In 1966, Aimard et al. described the arrest of the disease observed in a progressive case of MS using cy [11]. In subsequent years, the effects of the drug on patients with a progressive form of the disease (Table 1) were described [50][51][52][53][54][55]. In 1975, Drachman et al. observed no effects using 4-5 mg/kg of cy given for 10 successive days for the treatment of acute attacks (see Table 1) [56]. In 1977, Gonsette et al. described, in an open-label, uncontrolled study, the results obtained with cy in 201 relapsing/remitting patients [57] ( Table 1). It is interesting to note that more positive effects were observed in those patients with the shortest length of disease; on the contrary, more severely disabled patients did not benefit from cy treatment. In 1983, the first clinical trial with a rigorous design (randomized and controlled trial) was performed to evaluate the clinical effects of cy in MS patients. Hauser et al. [58] evaluated patients with progressive MS treated with a 2-to 3-week course of cy, intravenously 400-500 mg/day, to achieve leukopenia of 2000/mm 3 plus ACTH compared to a similar group treated with ACTH alone and to a group that received plasma exchange, ACTH, and oral cy. The results showed that 80% (16/20) of the cy-treated patients had improved or were stable at 1 year compared to only 20% (4/20) in the ACTH-treated group. The plasma exchange group showed an intermediate (50%) response.
Positive clinical results were observed on disability with a very low number of treatment failures. Moreover, analysis of the patient profiles demonstrated that the patients who were relatively young (35 years) and with a short disease duration (between 2 and 3 years) were the best clinical responders. It was also reported that 11/20 (55%) patients who were stable or who had improved at the 1-year follow-up experienced reprogression of their disease in the second or third year after treatment suggesting that a short period of treatment was not sufficient. In 1993, the Northeast Cooperative Treatment Group demonstrated that patients between 18 to 40 years old receiving cy boosters every other month for 2 years, did not show a reprogression of the disease and that there were no differences between the modified induction regimen and the previous published regimen to prevent or delay disease reprogression [59]. This study led the way for currently used protocols in which treatment is given as outpatient pulses similar to lupus nephritis treatment.
The Northeast Cooperative Study results were challenged by the results obtained with cy, on patients with progressive MS, from the Canadian Cooperative Multiple Sclerosis group study [60] and by the Kaiser study [61].
The first trial, a single blinded randomized and placebocontrolled multicentre study, included 168 progressive MS patients. The investigators did not find significant differences in time-to-treatment failure, comparing patients treated with intravenous cy and oral prednisone, patients treated with oral cy and oral prednisone on alternate days plus weekly plasma exchange and one further group of patients treated with oral placebo and sham plasma exchange. In the Kaiser study, 22 progressive MS patients received 400-500 mg of cy, IV, 5 days per week, until the leukocyte count fell below 4000/mm 3 , and were compared to 20 patients receiving folic acid in a randomized, single-blind study. The similar disease progression in the two groups provided evidence of a lack of substantial benefits of cy treatment.
The potential reasons for differences between these two studies and the 1983 NEJM study and the Northeast Cooperative Treatment Group study have been extensively debated in the literature [62][63][64].
In summary, the studies cited seem to delineate a role for cy in the treatment of patients with an inflammatory component of the disease (early, aggressive, and inflammatory MS). There is evidence of poor or no benefit when cy is administered in both primary and secondary progressive MS patients.

Recent Studies
In recent years, considering previous studies, most physicians have used cy to treat RR-MS, SP-MS, or rapidly deteriorating MS patients with intermittent intravenous (monthly or bimonthly) pulse therapy, alone or in combination with DMTs, and, less frequently, they have used an induction protocol (see Table 2). Of the several published studies (see Table 1), three must be highlighted because they were performed on patients with aggressive forms of the disease. Weinstock-Guttman et al. described the effects of cy (openlabel) in 17 consecutive patients with "fulminant MS" (defined as a deterioration of more than one and a half points on the EDSS for more than 3 months); after 24 months, 69% of patients were stable or had improved [65]. Gobbini et al. reported clinical stability in five patients with rapidly deteriorating relapsing-remitting MS who were treated monthly with cy for 6 months and then on alternate   After 1 year of treatment reduction in relapse rate and a stabilization of disability months. These results were confirmed by a rapid reduction in gadolinium-enhancing lesions, seen with monthly brain MRI scans [66]. Khan et al. reported an open-label study of intravenous cy given monthly to 14 patients with relapsingremitting MS who were rapidly deteriorating (defined as a greater than three-point increase in the EDSS score in the previous 12 months despite DMT therapy and intravenous prednisone) [67]. Some authors have used, accidentally or intentionally, high doses of cy. A report of a single patient with relapsing remitting MS who, on one occasion, accidentally received a dose of 3800 mg of the drug showed no evidence of clinical or MRI disease activity for the next 7 years [68]. An open-label study by Gladston and coworkers, with high doses of cy (200 mg per kg over 4 days), in 13 patients with treatment-refractory MS showed significant disease stability (no patients increased their baseline EDSS score by more than 1.0) and improvement in quality of life after 15 months [69]. Also, Krishman and coworkers described the effects of high doses of cy after 24 months in 21 patients with RR-MS with "active" MRI or clinical exacerbation in the year before or a worsening EDSS score of 1 point or higher compared to the preceding year using an immunoablative regimen of 50 mg/kg/day for 4 consecutive days followed by granulocyte colony stimulating factor Autoimmune Diseases 7 (GCS) 6 days after cy. They observed a reduction of the EDSS score and of the number of gadolinium-enhancing lesions at the end of follow-up [70]. Makhani and coworkers reviewed their multicentre experience with cy in the treatment of 17 children with MS. After 1 year of treatment they observed a reduction in relapse rate and a stabilization of disability scores [71].

Comparison Studies
Zipoli et al. compared the efficacy and safety of intravenous cy and mitoxantrone as second-line therapy in relapsingremitting or secondary-progressive MS patients. Mitoxantrone was administered at a dosage of 8 mg/m 2 monthly for 3 months, then every 3 months, until a dosage of 120 mg/m 2 was reached. Cy was administered at a dosage of 700 mg/m 2 monthly for 12 months, then bimonthly for a further 24 months. Seventy-five patients received mitoxantrone (31 RR, 44 SP) and 78 cy (15 RR, 63 SP).
The two groups differed only in terms of a significantly higher proportion of RR patients in the mitoxantrone group. After a mean follow-up of 3.6 years, the authors observed a lack of a significant difference in terms of time to the first relapse, whereas time to disease progression was slightly shorter in the group treated with mitoxantrone than in the cy group. After 12 months of treatment, active MRI scans were reduced by 69% in the mitoxantrone group and 63% in the cy group of patients. Discontinuation due to side effects was more frequent in cy patients, but the authors concluded that the overall tolerability of the two treatments was acceptable [72]. Gallo et al. treated fifty secondary progressive MS patients, who had lost one or more EDSS points in the prior two years, with cy (25 patients, mean disease duration 13.3 years; mean EDSS score at study entry: 5.7) or mitoxantrone (25 patients, mean disease duration: 11.5 years; mean EDSS score at study entry: 5.5). SPMS patients were treated for two years. The authors observed a significant reduction in both groups of relapse rate and disability progression. Subgroups of mitoxantrone-and cyresponding patients were characterized by a significantly shorter duration of the secondary progressive phase of the disease. In these subgroups, the improvement in the EDSS score at the end of therapy was higher than the remaining patients. The safety profiles of both drugs were acceptable; however, the authors also evaluated the cost of the two treatments and concluded that the cy-based therapy protocol was significantly less expensive [73].

Our Experience
Our group reported on the effectiveness of a combination of cy and beta-interferon in patients with rapidly progressive or "transitional" MS (characterized by frequent and severe attacks plus worsening on the disability status scale). We treated 10 such patients with monthly pulses of IV cy (500-1500 mg/m 2 ) to obtain a lymphopenia of between 600 and 900/mm 3 for 12 consecutive months and then at 2-month intervals for a further 6 months. We found a significant reduction in the number of relapses, disability and T2 MRI burden of disease with a stabilization of the disease for a mean of 36 months after the discontinuation of cy treatment [74,75]. We found that the treatment was safe and well tolerated. In the following years, we replicated this result in a cohort of thirty rapidly deteriorating MS patients treated for 24 months with cy (500-1500 mg/m 2 ) combined with INF-β [76]. In a randomised, multicentre trial of 59 patients with relapsing-remitting MS, who did not respond to interferon beta, Smith et al. observed that the combination of cy and interferon beta-1a reduced the clinical disease activity and gadolinium-enhancing MRI lesions in the brain [77]. In 2009, our group reported the effects of cy as initial induction therapy in "active" MS. Forty active relapsingremitting MS patients (>1 relapse in the prior 12 months and >1 Gd+ MRI lesion) were randomized in two groups of twenty to receive monthly cy, designed to induce a leucopenia below 1000× mm 3 , plus methylprednisolone (MP) 1 g for 12 months followed by IFN-β for a further 12 months versus IFN-β alone for 2 years. The annual relapse rate was reduced from 1.9 observed at baseline to 0.1 in the cy group versus 0.5 in the interferon group (P = .02) at Year 2; relapse-free patients at the second year were 80% of the cy group versus 40% of the IFN group (P = .024); and the percentage of patients without Gd+ MRI lesions at 24 months was 90% in the cy group versus 54% in the IFN group-2 (P = .04). No serious adverse events were observed during follow-up. This study supports the concept of using cy as an induction treatment for improving the impact of IFN over time [78]. A similar approach was used by Perumal et al. They used cy as the initial therapy in patients who had not received DMT therapy before. All patients had experienced at least two relapses in the year prior to therapy. Twenty-six patients received monthly intravenous cy for 6 months followed by initiation of immunomodulatory therapy. At year 1, the mean EDSS score, relapse rate, and Gd-enhancing per patient at baseline were reduced from 3.61 to 2.22, 3.42 to 0.77, and 3.55 to 0.33, respectively. These studies suggest that cy may be used as initial therapy in relapsing-remitting MS patients [79].

Cy in Autologous Haematopoietic Stem-Cell Transplantation (AHSCT) for MS
Intense immunosuppression using cy 1.5-4 g/m 2 total dose over 1-2 days, to mobilize peripheral blood hematopoietic stem cells (HSC), followed by a conditioning regimen, and then autologous haematopoietic stem-cell transplantation (AHSCT) has been evaluated as a possible new therapeutic tool in severe autoimmune disorders, after it was shown to be efficacious in animal models of immunomediated diseases [80]. The conditioning regimen, the second step of the procedure, could be carried out with several protocols, the most common protocol used is the BEAM regimen which includes 300 mg/m 2 carmustine at day-7, 200 mg/m 2 etoposide and 200 mg/m 2 cytarabine from day-6 to day-3, and 140 mg/m 2 melphalan at day-2. These drugs cross the blood-brain (a) IV induction therapy: 600 mg/m 2 CTX given on days 1, 2, 4, 6, 8 plus MP given daily for 8 days.
(b) IV pulse therapy with CYX/MP: CTX pulses starting at 800 mg/m 2 with dose augmentation in order to obtain a leukopenia of 3000/mm 3 and a lynphocitopenia of 800/mm 3 ; every 4 weeks x 18-24 months, every 2 months × 24. 1 g MP given with CTX.
(c) Pulse therapy with CTX at a fixed dose: CTX pulses given at 800-1000 mg/m 2 every 4-8 weeks for 12-24 months, given with or without MP.
(d) Combination therapy: IV pulse CTX therapy given concomitantly with beta-interferon or glatiramer acetate in nonresponders.
barrier and possibly hit the autoreactive clones that colonize the CNS. In North America and also in Europe, total body irradiation (TBI) plus cy is frequently used. Another conditioning regimen utilized in the reported studies is BCNU 300 mg/m 2 plus cy 150-200 mg/kg, an intermediate intensity regimen, which eliminates etoposide and ARA-C from the BEAM scheme. Less intense regimens, with a prevalence of immunosuppressive and minor myeloablative effects, such as cy 200 mg/kg plus alemtuzumab (Campath) or ATG, have been recently utilized, considering that they might provide a satisfactory clinical outcome with lower toxicity. The effects of AHSCT on the number of relapses are considerable; they are usually dramatically reduced; this anti-inflammatory activity is corroborated by the effects of AHSCT on MRI parameters, with the absence of gadolinium enhancing lesions on MRI inflammatory activity in most cases for a period of at least 1-3 years.
The main problem of AHSCT is the transplant-related mortality (TRM), which is around 3.3%. The mortality risk is mainly due to infections which can occur during the aplastic phase which follows the conditioning regimen and are related to the intensity of the therapies used [81][82][83][84][85].

How to Use Cy: What We Have Learned from Previous Studies and Our Experience
Over the last few years, the use of cy in MS has evolved towards the use of intermittent pulse therapy given monthly or bimonthly over a 1-to 3-year period administered intravenously with an adjusted dose to obtain a leucopoenia target or with a fixed dose ( Table 2, regimen (b) and (c) respectively). In our experience after the first dose, (usually we start with a dose of 800 mg/m 2 ), a dose modulation is made on the basis of the number of white blood cells observed at the nadir of leukopenia (after 12-14 days of administration). Our target is to keep the white blood cells below 3000 for mmc and lymphocytes below 800 for mmc. In our opinion, this target is related to a good balance between effectiveness and safety. Antiemetic drugs and iv steroids are usually administered and large amount of fluids are also administered during the same session of therapy. In treated patients, we further administer iv MESNA (with a dose equal to 20% of the dose of cy) before the cy booster and four hours after the administration of the drug. We also suggest that the patient should drink a lot (at least two litres of fluids) in the two days following treatment. MESNA is used therapeutically to reduce the incidence of hemorrhagic cystitis and hematuria when a patient receives cy. MESNA assists the neutralization of the urotoxic metabolites derived from the metabolism of cy by binding them through their sulfhydryl group and also increases urinary excretion of cysteine [86].

When to Use Cy: What We Have Learned from Previous Studies and Our Experience
Collectively, data from studies indicate that patients with rapidly worsening, treatment refractory, relapsing-remitting MS, or in an early secondary progressive phase of the disease might benefit from treatment with intravenous cy. We think that in selected patients the drug should be considered when there are reasons that do not allow use of fingolimod or natalizumab. Little or nothing appears to be effective in primary progressive forms of MS. In order to turn off inflammation from the beginning of the disease, an innovative approach could be (supported by the knowledge about MS pathogenesis), in selected patients, a short early treatment of 12-24 months with cy, followed by an immunomodulator drug. This approach could be, in our opinion, a reasonable treatment for a young patient, suffering from RR-MS with a short disease duration, who shows an "active" clinical disease, characterized by several relapses over a short period of time (for RR-MS patients) and/or active MRI disease (defined as either new T2 lesions or T1 gadolinium enhancing lesions). We can further consider the option to treat as early as possible, immediately in patients with a clinically isolated syndrome, with a higher T2 lesion load and gadolinium enhancing lesions. It is conceivable that the induction with a more potent agent (cy, mitoxantrone, natalizumab, alemtuzumab, fingolimod, and cladribine) followed by a less potent drug (interferons, GA) may "freeze" and stabilize the disease. The relatively short period of induction with cy allows us to use cumulative doses which are largely below the tolerated cumulative dose of 80-100 g/lifetime. Thus, if after 5-10 or more years patients tend to reprogress, there exists the option to treat with cy again for some time.

Toxicity
The toxic effects of cy in the treatment of neoplastic or immunomediated diseases are well known [87]. The most frequent serious adverse event is hemorrhagic cystitis [88,89]. Cases of bladder cancer have been observed in patients who have received long-term cy treatment [90]. For these, reasons long-term oral cy has been avoided in MS. Gonadal failure occurs in both men and women receiving alkylating agents such as cy. Most of the available data concerns the rate of ovarian failure in cancer survivors. However, in these patients, alkylating agents were used as part of a multidrug regimen and at different doses than for immunologic diseases. As concerns the use of cy in immunemediated disease, in a lupus nephritis trial, after 6-month courses then every 3 months of treatment for at least two more years, with a dose of 0.75 mg/m 2 , and then adjusting the dose based on the nadir, 23 out of 46 women (50%) developed amenorrhea [91]. In these studies, an age greater than 30 years and cumulative dose over 300 mg/kg were risk factors for persistent amenorrhea. The rate of amenorrhea in women with MS (approximately 40%-80% in large series) appears similar to that reported for rheumatic diseases [92]. Our data (unpublished) from more than 200 patients treated with the drug (alone or in combination with interferon-beta) showed a rate of irreversible ovarian failure of 18%, while transient amenorrhea was observed in 60% of fertile woman ( were normospermic (sperm count >20 mil/mL). All patients showed a decrease in spermatozoa motility (<50%) and teratospermia (atypical form >30%). All oligospermic and azospermic patients presented higher serum levels of FSH and LH [94]. After one year of follow-up, we observed a recovery of the spermatogenesis in 60% of these patients. An increased incidence of subsequent malignancies has been reported in non MS patients treated with cy. This event may occur a few years after cessation of therapy. The risk appears to increase as a function of total dose (care must be taken with cumulative lifetime doses exceeding 80-100 g) [95]. De Ridder et al. observed bladder cancer in five patients out of 70 (5.7%) treated with the drug and chronically catheterized. They suggested regular cystoscopy in these patients to allow early detection of bladder tumors [96]. Portaccio et al. assessed the safety and tolerability of cy "pulse" therapy, in 120 MS patients with progressive or very active MS who received intravenous monthly "pulses" of the drug for 12 months at the dosage of 700 mg/m 2 of body surface, then bimonthly for another 12 months. They evaluated the frequency and the severity of side effects, most commonly definitive amenorrhea (33.3% of fertile women), hypogammaglobulinemia (5.4%), and hemorrhagic cystitis (4.5%). Malignancies were diagnosed in four (3.6%) subjects, three of whom were previously treated with azathioprine [97]. As described above, Makhani and coworkers reviewed their multicentre experience with cy (cumulative dose between 1.60-72.70 g) in the treatment of 17 children with MS. The side effects observed were vomiting, transient alopecia, osteoporosis, amenorrhea, and a bladder carcinoma in one patient [71].
In our center, safety is monitored by obtaining blood and urine analyses every month and urine cytological examination every 3 months; bladder echography to evaluate an incomplete empting is obtained before and during treatment, ECG, chest radiography, echography of liver, spleen, kidney, bladder, uterus and lymph nodes, and mammography are performed every 12 months. Analysis of seminal fluid is obtained at the beginning and at the end of treatment. We suggest sperm cryopreservation or ovarian protection before cy treatment. We routinely administer a cumulative dose of about 30 g of the drug divided into monthly boosters over a period of one-two years. This treatment protocol is safe and well tolerated. It usually takes less than half of the cumulative recommended dose thus treatment can resume, if needed, in the course of the disease. Also, mitoxantrone cannot be administered safely for long periods. It appears that after 6-7 boosters, 70 mg/m 2 , mitoxantrone could increase the risk of cardiotoxicity and leukemia [98][99][100][101][102][103][104][105]. The use of natalizumab has raised some concerns about the risk of developing progressive multifocal leukoencephalopathy (PML). This risk is increased after 24-30 administrations and if patients had been previously treated with immunosuppressive agents [106][107][108]. However, the results observed from the patients included in the clinical trials and from the TOUCH prescribing program clearly indicate that the risk of PML during treatment with natalizumab is much lower than that of malignancies during treatment with cy (the risk ranged from 1/100.000 for the first year of treatment until 1/1000 after four years of treatment) [109,110]. Little is known about the safe cumulative dosage of the new oral agents such as cladribine and fingolimod. However, safety profiles of both drugs raise several concerns, even if, to date, there are no indications about long-term toxicity [110][111][112].

Conclusions and Future Investigations
Based on the literature and on our experience, we suggest that cy could be a therapeutic option in MS patients, especially if they have an active inflammatory component. On the contrary, in later stages of the disease or in patients with primary progressive MS the effects of the drug are very poor. Its ineffectiveness in later stages of the disease, when there is less inflammation and more degenerative processes, appears to be true also for other drugs currently used in the treatment of the disease.
Given that early inflammatory events appear to correlate with later disability, a major question is whether strong antiinflammatory drugs such as cy or other drugs currently used to treat the disease will have an impact on later degenerative changes if given early in the disease to halt inflammation. With a better understanding of the pathogenesis of the disease, the possibility to identify new biomarkers and the introduction of pharmacogenomics and genetics in MS it may be possible to identify responders and nonresponders and provide them with a tailored therapy. Without this information, cy is, in our opinion, a possible treatment option for people with MS.
From the approval of mitoxantrone for worsening forms of MS the question remains of the place of cy in this patient group.
Mitoxantrone is considered easier to administer than cy. However, because of cardiac toxicity and the observed high frequency of iatrogenic leukemia, it can only be given for a limited period and cannot be given again if patients begin to reprogress. Cy can be used as a retreatment drug given that the cumulative dose limit is higher than that of mitoxantrone. However, bladder toxicity can strongly limit further therapy. For this reason, in our opinion, the use of cy or mitoxantrone depends, on the clinical condition of the organs suitable to be damaged by each drug (bladder and heart, respectively) and by the diligent monitoring over time of the patient's clinical conditions to identify and promptly treat any complications rather than on the cumulative dose. Furthermore, follow-up needs to be extended for many years following the treatment period to identify unknown long term side effects and to detect and treat, as soon as possible any that occur.
Sequential use of these agents has been carried out by some investigators, but toxicity profiles are unknown at this time.
We conclude that therapy with cy given in pulse therapy or in some instances as an acute induction regimen has an ameliorating effect on the disease process. It could be used in selected MS patient groups, and in "good hands" it could be used as induction regimen therapy in larger patient populations.