- Email: [email protected]
Predictors of progression in primary progressive multiple sclerosis in a large Turkish cohort
Multiple Sclerosis and Related Disorders 38 (2020) 101520
Contents lists available at ScienceDirect
Multiple Sclerosis and Related Disorders journal homepage: www.elsevier.com/locate/msard
Clinical trial
Predictors of progression in primary progressive multiple sclerosis in a large Turkish cohort
T
⁎
Atay Vurala,b, , Eda Derlea,c, Güliz Sayat-Güreld, Rana Karabudakd, Aslı Tuncera,d a
Institute of Neurological Sciences and Psychiatry, Hacettepe University, Ankara, Turkey Department of Neurology, School of Medicine, Koç University, İstanbul, Turkey c Department of Neurology, Başkent University, Ankara, Turkey d Department of Neurology, Faculty of Medicine, Hacettepe University, Ankara, Turkey b
A R T I C LE I N FO
A B S T R A C T
Keywords: Primary progressive multiple sclerosis Prognostic factors Disease activity Natural history Magnetic resonance imaging
Background: Studies on the predictors of progression for primary progressive multiple sclerosis (PPMS) are limited and there is no information in the literature for populations outside Europe and North America. In this study, we aimed to identify predictors of progression in a large Turkish PPMS cohort. Methods: We analyzed a cohort of 157 PPMS patients to investigate the effect of age of onset, gender, onset symptoms, presence or absence of relapses, and baseline gadolinium-enhancing lesions on the rate of progression to EDSS6 by using Kaplan-Meier analysis and multivariate Cox regression. Results: Older age of onset and presence of spinal motor symptoms at onset were associated with a shorter time to EDSS6 and presence of supratentorial signs at onset was associated with a longer time to EDSS6 according to Kaplan-Meier analysis. These factors remained significant after multivariate Cox-regression analysis. Clinical relapses were present in 22.3% and gadolinium-enhancing lesions on baseline MRI were present in 28% of patients, but these factors were not predictive of time to EDSS6. Conclusion: We identified age of onset and symptom at onset as predictors of progression in Turkish PPMS patients. Presence of clinical relapses or baseline gadolinium-enhancing lesions did not affect PPMS progression rate.
1. Introduction Multiple sclerosis (MS) is a heterogeneous disease with a variety of disease phenotypes and different prognosis among individuals. Due to this variability it is a challenge for clinicians to predict prognosis and provide optimal treatment on an individual basis. Prediction of prognosis from individual characteristics is also important for the design of clinical trials. Primary progressive MS (PPMS) has different pathological and clinical characteristics compared to relapsing-remitting multiple sclerosis (RRMS) and effective treatment options are not available for PPMS, unlike RRMS. All the previous clinical trials had been unsuccessful, except the recent trial of ocrelizumab which showed a modest beneficial effect (Sorensen and Blinkenberg, 2016). Analysis of these trials revealed that, one of the reasons of failure may be the overall slow rate of disease progression in recruited patients. Accordingly, PROMise study was terminated early because of slow disease progression in PPMS patients during the study period (Wolinsky et al.,
⁎
2007). In a previous study, it was reported that 25% of PPMS patients had reached EDSS 6.0 in 5 years, but after 17 years 25% still had not reached this milestone (Harding et al., 2015). In another study, it was found that 9% of the patients with PPMS didn't reach an EDSS score of 3 for at least ten years (Koch et al., 2009). These factors underline the importance of the longitudinal studies which focus on the effect of possible prognostic factors. Clinical and radiological markers of disease activity have long been thought to be disease modifiers in PPMS, as in relapsing-remitting MS. Clinical relapses were not found to be associated with prognosis in previous studies (Harding et al., 2015; Andersson et al., 1999; Confavreux et al., 2000) and radiological markers are not investigated in long-term longitudinal studies yet. In this study, we aimed to investigate the prognostic value clinical variables including the age of onset, gender and localization of the onset symptom together with presence of relapses and gadolinium-enhancing lesions on baseline magnetic resonance imaging (MRI) in a cohort of Turkish PPMS patients.
Corresponding author at: Department of Neurology, School of Medicine, Koç University, 34010 Zeytinburnu, Istanbul, Turkey E-mail address: [email protected] (A. Vural).
https://doi.org/10.1016/j.msard.2019.101520 Received 25 May 2019; Received in revised form 7 November 2019; Accepted 10 November 2019 2211-0348/ © 2019 Elsevier B.V. All rights reserved.
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
Fig. 1. Study design and workflow for data extraction and validation. PPMS: primary progressive multiple sclerosis; SAPMS: single-attack progressive multiple sclerosis; SPMS: secondary progressive multiple sclerosis; RRMS: relapsing remitting multiple sclerosis; EDSS: expanded disability scaling system; PI: principle investigator.
2. Methods
superimposed relapses (classified as primary relapsing MS) were included in the study cohort.
2.1. Study setting 2.3. Data extraction and validation Study was conducted on a hospital-based cohort. Specialized MS clinic in Hacettepe University has been active since 1995 and serves as one of the main referral centers for MS patients from all regions of Turkey. All patients were followed by one of the two MS specialists (RK or AT) since 1995. Detailed information on disease onset, demographic data, detailed neurological examination of the patient and laboratory and neuroimaging features are recorded in the baseline visit. In each follow-up visit, clinical relapses and disease course since last follow-up, current symptoms, medications, detailed neurological examination and neuroimaging findings are recorded to patient charts prospectively. In 2008, demographic data of all the previous patients who were diagnosed with MS in our center were recorded in the iMED local electronic database. In 2013, this database was updated with extensive clinical and paraclinical data and all MS patients seen in the outpatient clinic and neurology ward are recorded prospectively to the database since then. This study was approved by Hacettepe University Ethical Committee.
Data extraction workflow is as follows (Fig. 1): Patients seen between 1995 and 2015 were searched through database. 222 patients noted as “PPMS” or “PPMS-like” were detected. Of these, 30 patients were concluded as “not PPMS” (SAPMS, SPMS, RRMS, other). Next, the iMed database and all available digital and manually written patient charts were independently reviewed by two neurologists specializing on MS (AV and ED) for clinical and neuroimaging information. During this step, unclear information was clarified by phone calls to patients (n = 16). EDSS scores were derived directly from chart or determined from neurological examination. Conflicts on age of onset, type of starting symptoms and age of EDSS 6 were resolved with the primary physician of the patient (either AT or RK). In the last step, 34 patients were excluded because of inadequate information on disease onset, diagnosis or follow-up. One patient with PPMS was excluded because he had a concomitant disabling disease. Presenting symptoms were grouped under “spinal cord” referring to bilateral motor symptoms with upper motor signs, “brainstem/cerebellum” or “optic nerves” when symptoms of these regions were detected. “Supratentorial” symptoms were defined as cognitive or unilateral sensory/motor symptoms in the presence of compatible cerebral lesions on MRI and absence of bilateral motor and brainstem/cerebellar symptoms.
2.2. Patient selection Patients with a diagnosis of PPMS were identified from database. Poser or McDonald criteria were used for diagnosis depending on the year in which the patient was diagnosed. Progression of disease for at least one year from onset without any previous relapses was necessary for a patient to be classified as PPMS. Patients with a suspicious history of relapse before disease onset were excluded. Patients with
2.4. Collection of the neuroimaging data All patients were evaluated for the presence of gadolinium-enhancing lesions in brain and cervical spinal MRI in the baseline visit. 2
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
Table 1 Characteristics of patient population and comparison of patients with and without relapses and baseline gadolinium-enhancement.
Gender, F:M (n) Age of onset, y mean ± SD median, IQR Age of onset groups, n (%) 20 to <30 30 to <40 40 to <50 50< Localization of the presenting symptom, n (%) Spinal cord Brainstem/cerebellum Supratentorial Optic nerves Time to diagnosis, y mean ± SD median, IQR EDSS at diagnosis mean ± SD median, IQR Oligoclonal bandsa, n (%) Treatment status with DMT Follow-up, y mean ± SD median, IQR Age of EDSS6, y mean, 95% CI median, 95% CI Time to EDSS 6, y mean, 95% CI median, 95% CI
Total (n = 157)
PPMS with relapses (n = 35)
PPMS without relapses (n = 122)
p value
PPMS with baseline gadolinium-enhancement (n = 44)
PPMS without baseline gadolinium -enhancement (n = 113)
p value
1.5 (94/63)
1.9 (23/12)
1.4 (71/51)
0.443
1.75 (28/16)
1.4 (66/47)
0.590
34.8 ± 9.7 35.6, 27.7–41.1
32.7 ± 9.1 33, 25–37
35.4 ± 9.8 36, 27.8–42
0.139 0.104
33.7 ± 8.3 34.4, 27.7–38.2
35.2 ± 10.1 36, 26.9–42.4
0.386 0.327
122 (77.7) 44 (28) 24 (15.3) 3 (1.9)
25 (71.4) 14 (40) 3 (8.6) 2 (5.7)
97 (79.5) 30 (24.6) 21 (17.2) 1 (0.8)
0.358 0.089 0.290 0.125
32 (72.7) 20 (45.5) 5 (11.4) 3 (6.8)
90 (79.6) 24 (21.2) 19 (16.8) 0 (0)
0.395 0.005 0.467 0.021
3.1 ± 3.6 2, 1–3.9
1.7 ± 1.1 1, 1–2
3.6 ± 4 2, 1–4.9
<0.001 0.001
2.1 ± 2.2 1, 1–2
3.6 ± 4 2, 1–4.4
0.004 0.002
3.2 ± 1.3 3, 2–4 58/80 (72.5%)
2.6 ± 1.3 2.5, 2–3.5 8 (53.3) 22 (62.9)
3.4 ± 1.3 3.5, 2.5–4 50 (76.9) 88 (72.1)
0.003 0.002 0.105 0.302
2.8 ± 1.2 2.5, 2–3.5 18 (75) 27 (61.4)
3.3 ± 1.3 3.5, 2.5–4 40 (71.4) 83 (73.5)
0.029 0.017 0.792 0.174
10.2 ± 6.6 8.5, 5–14
9.8 ± 6.5 7.7, 4.8–13.9
10.4 ± 6.7 8.7, 5–14.3
0.633 0.517
8.5 ± 5 7.1, 4.2–12.5
10.9 ± 7 9.2, 5.1–15.3
0.018 0.065
42.8, 40.9–44.7 44, 40.1–47.9
42.2, 38.5–45.9 41, 37.5–44.5
43, 40.8–45.2 45, 42.3–47.7
0.633*
41.1, 37.9–44.4 40, 36.6–43.4
43.4, 41.1–45.7 45, 43.4–46.6
0.172b
9.9, 8.7–11 8, 6.7–9.3
10.7, 8.1–13.4 9, 6.7–11.2
9.5, 8.3–10.6 8, 6.6–9.4
0.459*
9.5, 7.3–11.8 7, 4.5–9.5
10, 8.7–11.3 9, 7.5–10.5
0.576b
54 (34.4) 53 (33.8) 42 (26.8) 8 (5.1)
F: female; M: male; y: years; SD: standard deviation; IQR: inter quartile range; DMT=disease-modifying therapy, CI: confidence interval. a Data for cerebrospinal fluid oligoclonal band status was available from 80 patients. b Long-rank test.
Fig. 2. Kaplan–Meier survival curves showing time to expanded disability status scale 6 (EDSS6) by age at onset (A); presence or absence of spinal symptoms at onset (B); and presence or absence of supratentorial symptoms at onset (C).
Baseline MRI is defined as the first available gadolinium-enhancing brain and spinal cord MRI of a patient. MRI scans were obtained in the study center, unless another gadolinium-enhanced MRI of adequate quality was obtained previously in another center. MRI studies of the MS patients are reported routinely by a neuroradiologist in our center and images are also reviewed during patient visit by the MS specialist in charge and recorded to the patient chart. Additionally, if a patient had MRI done in a different center the images were interpreted by one of the MS specialists and this information was also recorded in the patient's
chart and included in our study. 2.5. Statistical analysis Clinical features including gender, localization of the first symptom, oligoclonal band positivity and treatment status with disease-modifying therapies (DMTs) were compared with chi-square test. Age of onset, time to diagnosis, EDSS at diagnosis and follow-up duration were compared with independent T-test and Mann-Whitney U test. Median 3
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
Table 2 Kaplan–Meier and multivariate Cox regression analysis results of potential predictors of time to EDSS6. Factor Gender Female Male Age at onset <30 30 to <40 40 to <50 50 and older Onset symptom localization Spinal Present Absent Brainstem/ Cerebellum Present Absent Supratentorial Present Absent Clinical relapses Present Absent Gadolinium-enhancement at baseline Present Absent
Total, n
Reaching EDSS6, n (%)
Median time to EDSS6 (95% CI)
p valuea
HR (95% CI) for time to EDSS6
p value
94 63
64 (68.1) 41 (65.1)
8.0 (6.5–9.5) 9.0 (7.8–10.2)
0.587
–
ns
54 53 42 8
36 (66.7) 37 (69.8) 25 (59.5) 7 (87.5)
9.0 (7.4–10.6) 10.0 (8–12) 7.0 (5.8–8.2) 4.0 (1.5–6.5)
0.008
1.03 (1.006–1.053)b
0.012
122 35
89 (73) 16 (45.7)
8.0 (6.6–9.4) 13.0 (10.8–15.2)
0.003
2.13 (1.24–3.63)
0.006
44 113
23 (52.3) 82 (72.6)
9.0 (5.2–12.8) 8.0 (6.3–9.7)
0.085
–
ns
24 133
10 (41.7) 95 (71.4)
14.0 (12.4–15.6) 8.0 (6.8–9.3)
0.003
0.40(0.21–0.78)
0.007
35 122
23 (65.7) 82 (67.2)
9 (6.7–11.2) 8 (6.5–9.5)
0.459
–
ns
44 113
27 (61.4) 78 (69)
7 (4.5–9.5) 9 (7.5–10.5)
0.576
ns
a
Long-rank test. increased risk per year. EDSS: expanded disability status scale; HR: hazard ratio; CI: confidence interval; ns: not significant. Statistically significant p values are written in bold. b
time to EDSS6 for younger age of onset (p = 0.009) (Fig. 2A). Patients with spinal symptoms at onset had a faster progression rate (p = 0.003) (Fig. 2B). Presence of supratentorial symptoms at onset, including cognitive symptoms or sensory symptoms in the absence of bilateral motor symptoms or urinary symptoms, was associated with a longer time to EDSS6 (p = 0.002) (Fig. 2C). Gender, treatment with DMTs, presence of relapses or baseline gadolinium-enhancing lesions did not influence time to EDSS6. To study the independent effect of these factors on time to EDSS6 we introduced presence or absence of spinal and supratentorial symptoms at onset, and presence or absence of relapses and baseline gadolinium-enhancing lesions together with gender and age of onset (as continuous variable) into the Cox regression model. Older age of onset (HR value per 1-year increase=1.03, 95% CI:1.006–1.053, p = 0.012) and presence of spinal symptoms at onset (HR=2.13, 95% CI:1.24–3.63, p = 0.006) were independently associated with a shorter time to EDSS6. Presence of supratentorial symptoms at onset was independently associated with a longer time to EDSS6 (HR=0.373, 95% CI:0.186–0.745, p = 0.005) (Table 2). Presence of relapses or gadolinium-enhancing lesions on baseline MRI did not influence neither time to EDSS6 nor age at EDSS6. Regarding the age at EDSS6, younger age of onset category was associated with also a younger age at EDSS6 with both Kaplan-Meier analysis (p<0.001) and Cox regression model (HR=0.87, 95% CI: 0.84–0.89). None of the other factors examined were associated with age at EDSS6.
age of EDSS6 and median time to EDSS6 for different gender, age at onset, location of first symptom, treatment status with DMTs, baseline gadolinium-enhancing lesion status and clinical relapse status categories were calculated by Kaplan-Meier survival analysis. In the model generated for multivariate Cox regression analysis, same factors analyzed by Kaplan-Meier analysis were entered as variables, with age as continuous and others as categorical variables. To compare the mean age of onset of our population to European populations, adjusted-mean age of onset was calculated.
3. Results Study population consisted of 157 PPMS patients with a median (IQR) follow-up time of 8.5 (5.0–14.0) years. Median age of onset was 35.6 years (95% CI:27.7–41.1), median age at EDSS6 was 44 years (95% CI:40.1–47.9) and median time to EDSS6 was 8 years (95% CI:6.7–9.3). Presenting symptoms localized to spinal cord in 77.7% (n = 122), brainstem/cerebellum in 28% (n = 44), supratentorium in 15.3% (n = 24), optic nerves in 1.9% (n = 3). Among patients presenting with supratentorial symptoms, 11 had unilateral motor (7%), nine had unilateral sensory (6%) and four had combined unilateral motor and sensory symptoms in the absence of cervical lesions (3%); eight patients (5%) presented with cognitive symptoms. Other clinical characteristics of the study population is shown in Table 1. At least one clinical relapse was present in 35 cases (22.3%). Eight of these patients had two relapses during follow-up and others had one relapse. 31 (88.5%) patients had the relapse in the first five years after onset. Clinical characteristics were similar between relapsing and nonrelapsing patients, except that relapsing patients had a shorter time to diagnosis and lower EDSS at diagnosis. Gadolinium-enhancing lesions were present in the baseline MRI of 44 (28%) patients. Again, clinical characteristics were similar between patients with and without baseline gadolinium-enhancing lesions, except for a shorter time to diagnosis and a lower EDSS at diagnosis for patients with baseline gadolinium-enhancing lesions. Kaplan-Meier survival analysis revealed that time to EDSS6 was different for patients in different age of onset categories, with a longer
4. Discussion Our main findings can be summarized as: Younger age of onset is associated with a longer time to EDSS6 in PPMS. Presence of spinal symptoms at onset is associated with a faster progression rate, whereas presence of supratentorial symptoms is predictive of longer time to EDSS6 in PPMS. Having a relapsing course or gadolinium-enhancing lesion on baseline MRI does not affect disease progression rate in PPMS. Median time to EDSS6 in our study population (8 years) is within the same range with the previously reported values (7.1–14) (Table 3), 4
Multiple Sclerosis and Related Disorders 38 (2020) 101520
5.8 3
57.3 45
9.6 9
Yes Yes
No No
No Yes
No No
indicating a similar disease progression rate in our population with the others (Harding et al., 2015; Koch et al., 2009; Andersson et al., 1999; Confavreux et al., 2000; Cottrell et al., 1999; Debouverie et al., 2007; Kelly et al., 2013; Koch et al., 2015; Leray et al., 2010; Raghavan et al., 2015; Tutuncu et al., 2013). Median age of onset was younger in our cohort compared to the previous natural history studies performed in Europe and North America (38.5–44.4) (Table 3). This result can stem from the hospital-based design of our study, which may have created a referral bias towards a more active and younger-onset cases. Alternatively, Turkish population has a different age distribution compared to European and North American populations, with a higher ratio of younger individuals and lower ratio for elder, and this may have a significant effect on the average of milestone ages. Effect of relapses, gender, age of onset, and symptom of onset on time to EDSS6 in PPMS patients have been studied in several studies (Harding et al., 2015; Koch et al., 2009; Andersson et al., 1999; Confavreux et al., 2000; Cottrell et al., 1999; Debouverie et al., 2007; Koch et al., 2015; Leray et al., 2010; Stellmann et al., 2014; Confavreux et al., 2003) (Table 3). Male gender had been associated with faster progression rate in one study (Confavreux et al., 2003), however this finding was not replicated in others. Association of lower age of onset with longer time to EDSS6 in PPMS was shown in three other studies (Harding et al., 2015; Koch et al., 2009; Koch et al., 2015) and also for SPMS (Paz Soldan et al., 2015). The relationship with age of onset and rate of disease progression in the progressive phase of MS indicates that neurodegeneration at that stage is at least partly an agedependent process. However, age at onset cannot be taken as a positive prognostic factor, as it is also associated with a younger age at EDSS6 as shown in our study and previous studies (Koch et al., 2009; Koch et al., 2015). Studies on the relation of symptom of onset to pace of disease progression yielded varying results. In earlier studies, symptom type was not associated with prognosis (Cottrell et al., 1999; Debouverie et al., 2007; Leray et al., 2010; Confavreux et al., 2003). However, in two recent studies, absence of bilateral motor symptoms and presence of sensory symptoms were associated with slower disease progression to EDSS6 (Koch et al., 2009; Koch et al., 2015). In our study, onset with spinal cord symptoms which was defined as bilateral motor symptoms, was associated with shorter time to EDSS6 and onset with supratentorial symptoms was associated with longer time to EDSS6. Altogether, these findings suggest that disease onset with symptoms other than bilateral motor and brainstem/cerebellar symptoms may be associated with slower disease progression. One recent neuroimaging study reported that in PPMS patients spinal cord volume (SCV) loss progressed independent of total brain volume and T2 lesion volume change and in PPMS, and SCV was the only MRI measurement associated with EDSS increase over time, as opposed to RRMS and SPMS (Tsagkas et al., 2019). Therefore, there may be a subgroup of PPMS patients in whom the pathology is localized predominantly to supratentorium, in relation with a slower disease progression rate. This hypothesis should be further investigated by a prospective MRI study. Gadolinium-enhancing lesions were present at baseline MRI in 28% of our patients. A few other studies reported the frequency and treatment-related prognostic value of gadolinium-enhancing lesions in PPMS. In a study which was performed with triple dose gadolinium, out of 45 patients with a mean disease duration of 3.3 years, 42% had gadolinium-enhancing lesions (Ingle et al., 2005). In the PROMise trial, gadolinium-enhancing lesions were detected in 14% of all baseline MRI scans from 943 patients (Wolinsky et al., 2007). In the OLYMPUS trial, 24.5% of 439 PPMS patients had gadolinium-enhancing lesions at baseline MRI and these patients had a better response to rituximab in this study (Hawker et al., 2009). In the ORATORIO trial, 27.2% of 732 PPMS patients had gadolinium-enhancing lesions at baseline MRI. After subgroup analysis, the authors concluded that the study was underpowered to show the potential of the gadolinium-enhancing lesions as a treatment-effect–modifying factor (Montalban et al., 2017). These
2015(3) 2019
y: years; F: female; M: male; EDSS: expanded disability status scale; NS: not specified.
2.9 2 43.9 35.6 1.1 1.49 13.1 8.5 27.8 22.3 234 157
No – No – – – – – No No No No Yes No Yes No No No Yes No No No No No No No No No Yes No Yes No 10.3 8 7.1 9.9 14 10 9 – 5.6 – – – – – – 4
San Francisco, USA Ontario, Canada Lyon, France Lorraine, France British Columbia, Canada Rennes, France Calgary, Canada Hamburg & Munich, Germany Cardiff, Wales Ankara, Turkey 1999(5) 1999(7) 2000(6) 2007(8) 2009(4) 2010(11) 2015(10) 2015(14)
95 216 282 359 552 445 500 597
12.6 NS 38.6 29.8 0% NS NS NS
13.9 23.7 – 13.6 12.9 – 14 4.4
1.44 1.3 – 1.36 1.13 1.41 1.12 1.17
40.8 38.5 – 42.7 41.8 38.5 44 40.4
– 9.8 – – 7.7 – – –
– – 53 – 58.6 52.1 55 –
Relapses Symptoms Sex Age
EDSS at diagnosis (y) Time to diagnosis (y) Age of onset (y) F:M Follow-up (y) PRMS (%) n Location Year (Reference)
Table 3 Previous studies providing data on predictors of progression for primary progressive multiple sclerosis.
Age of EDSS=6 (y)
Time to EDSS=6 (y)
Prognostic factors (time to EDSS6)
A. Vural, et al.
5
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
results show that, gadolinium-enhancing lesions are not rare in PPMS and they may have a prognostic value regarding treatment response, however this topic requires further investigation. There are several limitations of our study: One of the main limitations is the retrospective nature of the study, which is inevitable for such long-term studies. Also, although our study population was large enough to show the effect of age of onset and onset symptoms on progression rate, we cannot rule out that larger studies may show a relationship between baseline gadolinium-enhancement and prognosis. In conclusion, our study confirms the association between the age of onset and symptom at onset with time to EDSS6. Population-based natural history studies are necessary to investigate further the characteristics of PPMS in different populations.
Multiple sclerosis with a progressive course from onset in Lorraine-eastern France. J. Neurol. 254 (10), 1370–1375. Harding, K.E., Wardle, M., Moore, P., Tomassini, V., Pickersgill, T., Ben-Shlomo, Y., et al., 2015. Modelling the natural history of primary progressive multiple sclerosis. J. Neurol. Neurosurg. Psychiatry 86 (1), 13–19. Hawker, K., O'Connor, P., Freedman, M.S., Calabresi, P.A., Antel, J., Simon, J., et al., 2009. Rituximab in patients with primary progressive multiple sclerosis: results of a randomized double-blind placebo-controlled multicenter trial. Ann. Neurol 66 (4), 460–471. Ingle, G.T., Sastre-Garriga, J., Miller, D.H., Thompson, A.J, 2005. Is inflammation important in early PPMS? a longitudinal MRI study. J. Neurol. Neurosurg. Psychiatry 76 (9), 1255–1258. Kelly, S.B., Kinsella, K., Duggan, M., Tubridy, N., McGuigan, C., Hutchinson, M, 2013. A proposed modification to the Mcdonald 2010 criteria for the diagnosis of primary progressive multiple sclerosis. Mult. Scler. 19 (8), 1095–1100. Koch, M., Kingwell, E., Rieckmann, P., Tremlett, H, 2009. The natural history of primary progressive multiple sclerosis. Neurology 73 (23), 1996–2002. Koch, M.W., Greenfield, J., Javizian, O., Deighton, S., Wall, W., Metz, L.M, 2015. The natural history of early versus late disability accumulation in primary progressive MS. J. Neurol. Neurosurg. Psychiatry 86 (6), 615–621. Leray, E., Yaouanq, J., Le Page, E., Coustans, M., Laplaud, D., Oger, J., et al., 2010. Evidence for a two-stage disability progression in multiple sclerosis. Brain 133 (Pt 7), 1900–1913. Montalban, X., Hauser, S.L., Kappos, L., Arnold, D.L., Bar-Or, A., Comi, G., et al., 2017. Ocrelizumab versus placebo in primary progressive multiple sclerosis. N. Engl. J. Med. 376 (3), 209–220. Paz Soldan, M.M., Novotna, M., Abou Zeid, N., Kale, N., Tutuncu, M., Crusan, D.J., et al., 2015. Relapses and disability accumulation in progressive multiple sclerosis. Neurology 84 (1), 81–88. Raghavan, K., Healy, B.C., Carruthers, R.L., Chitnis, T, 2015. Progression rates and sample size estimates for PPMS based on the climb study population. Mult. Scler. 21 (2), 180–188. Sorensen, P.S., Blinkenberg, M, 2016. The potential role for ocrelizumab in the treatment of multiple sclerosis: current evidence and future prospects. Ther. Adv. Neurol. Disord. 9 (1), 44–52. Stellmann, J.P., Neuhaus, A., Lederer, C., Daumer, M., Heesen, C, 2014. Validating predictors of disease progression in a large cohort of primary-progressive multiple sclerosis based on a systematic literature review. PLoS ONE 9 (3), e92761. Tsagkas, C., Magon, S., Gaetano, L., Pezold, S., Naegelin, Y., Amann, M., et al., 2019. Preferential spinal cord volume loss in primary progressive multiple sclerosis. Mult. Scler. 25 (7), 947–957. Tutuncu, M., Tang, J., Zeid, N.A., Kale, N., Crusan, D.J., Atkinson, E.J., et al., 2013. Onset of progressive phase is an age-dependent clinical milestone in multiple sclerosis. Mult. Scler. 19 (2), 188–198. Wolinsky, J.S., Narayana, P.A., O'Connor, P., Coyle, P.K., Ford, C., Johnson, K., et al., 2007. Glatiramer acetate in primary progressive multiple sclerosis: results of a multinational, multicenter, double-blind, placebo-controlled trial. Ann. Neurol. 61 (1), 14–24.
Declaration of Competing Interest Atay Vural, Eda Derle, Güliz Sayat-Gürel, Rana Karabudak and Aslı Tuncer does not have any conflict of interest to report related to the study. Acknowledgements We are thankful to our patients and their care-givers for participation in the study. This study is dedicated to the memory of our beloved colleague and dear friend Güliz Sayat-Gürel. References Andersson, P.B., Waubant, E., Gee, L., Goodkin, D.E, 1999. Multiple sclerosis that is progressive from the time of onset: clinical characteristics and progression of disability. Arch. Neurol. 56 (9), 1138–1142. Confavreux, C., Vukusic, S., Adeleine, P, 2003. Early clinical predictors and progression of irreversible disability in multiple sclerosis: an amnesic process. Brain 126 (Pt 4), 770–782. Confavreux, C., Vukusic, S., Moreau, T., Adeleine, P, 2000. Relapses and progression of disability in multiple sclerosis. N. Engl. J. Med. 343 (20), 1430–1438. Cottrell, D.A., Kremenchutzky, M., Rice, G.P., Koopman, W.J., Hader, W., Baskerville, J., et al., 1999. The natural history of multiple sclerosis: a geographically based study. 5. the clinical features and natural history of primary progressive multiple sclerosis. Brain 122 (Pt 4), 625–639. Debouverie, M., Louis, S., Pittion-Vouyovitch, S., Roederer, T., Vespignani, H, 2007.
6
Contents lists available at ScienceDirect
Multiple Sclerosis and Related Disorders journal homepage: www.elsevier.com/locate/msard
Clinical trial
Predictors of progression in primary progressive multiple sclerosis in a large Turkish cohort
T
⁎
Atay Vurala,b, , Eda Derlea,c, Güliz Sayat-Güreld, Rana Karabudakd, Aslı Tuncera,d a
Institute of Neurological Sciences and Psychiatry, Hacettepe University, Ankara, Turkey Department of Neurology, School of Medicine, Koç University, İstanbul, Turkey c Department of Neurology, Başkent University, Ankara, Turkey d Department of Neurology, Faculty of Medicine, Hacettepe University, Ankara, Turkey b
A R T I C LE I N FO
A B S T R A C T
Keywords: Primary progressive multiple sclerosis Prognostic factors Disease activity Natural history Magnetic resonance imaging
Background: Studies on the predictors of progression for primary progressive multiple sclerosis (PPMS) are limited and there is no information in the literature for populations outside Europe and North America. In this study, we aimed to identify predictors of progression in a large Turkish PPMS cohort. Methods: We analyzed a cohort of 157 PPMS patients to investigate the effect of age of onset, gender, onset symptoms, presence or absence of relapses, and baseline gadolinium-enhancing lesions on the rate of progression to EDSS6 by using Kaplan-Meier analysis and multivariate Cox regression. Results: Older age of onset and presence of spinal motor symptoms at onset were associated with a shorter time to EDSS6 and presence of supratentorial signs at onset was associated with a longer time to EDSS6 according to Kaplan-Meier analysis. These factors remained significant after multivariate Cox-regression analysis. Clinical relapses were present in 22.3% and gadolinium-enhancing lesions on baseline MRI were present in 28% of patients, but these factors were not predictive of time to EDSS6. Conclusion: We identified age of onset and symptom at onset as predictors of progression in Turkish PPMS patients. Presence of clinical relapses or baseline gadolinium-enhancing lesions did not affect PPMS progression rate.
1. Introduction Multiple sclerosis (MS) is a heterogeneous disease with a variety of disease phenotypes and different prognosis among individuals. Due to this variability it is a challenge for clinicians to predict prognosis and provide optimal treatment on an individual basis. Prediction of prognosis from individual characteristics is also important for the design of clinical trials. Primary progressive MS (PPMS) has different pathological and clinical characteristics compared to relapsing-remitting multiple sclerosis (RRMS) and effective treatment options are not available for PPMS, unlike RRMS. All the previous clinical trials had been unsuccessful, except the recent trial of ocrelizumab which showed a modest beneficial effect (Sorensen and Blinkenberg, 2016). Analysis of these trials revealed that, one of the reasons of failure may be the overall slow rate of disease progression in recruited patients. Accordingly, PROMise study was terminated early because of slow disease progression in PPMS patients during the study period (Wolinsky et al.,
⁎
2007). In a previous study, it was reported that 25% of PPMS patients had reached EDSS 6.0 in 5 years, but after 17 years 25% still had not reached this milestone (Harding et al., 2015). In another study, it was found that 9% of the patients with PPMS didn't reach an EDSS score of 3 for at least ten years (Koch et al., 2009). These factors underline the importance of the longitudinal studies which focus on the effect of possible prognostic factors. Clinical and radiological markers of disease activity have long been thought to be disease modifiers in PPMS, as in relapsing-remitting MS. Clinical relapses were not found to be associated with prognosis in previous studies (Harding et al., 2015; Andersson et al., 1999; Confavreux et al., 2000) and radiological markers are not investigated in long-term longitudinal studies yet. In this study, we aimed to investigate the prognostic value clinical variables including the age of onset, gender and localization of the onset symptom together with presence of relapses and gadolinium-enhancing lesions on baseline magnetic resonance imaging (MRI) in a cohort of Turkish PPMS patients.
Corresponding author at: Department of Neurology, School of Medicine, Koç University, 34010 Zeytinburnu, Istanbul, Turkey E-mail address: [email protected] (A. Vural).
https://doi.org/10.1016/j.msard.2019.101520 Received 25 May 2019; Received in revised form 7 November 2019; Accepted 10 November 2019 2211-0348/ © 2019 Elsevier B.V. All rights reserved.
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
Fig. 1. Study design and workflow for data extraction and validation. PPMS: primary progressive multiple sclerosis; SAPMS: single-attack progressive multiple sclerosis; SPMS: secondary progressive multiple sclerosis; RRMS: relapsing remitting multiple sclerosis; EDSS: expanded disability scaling system; PI: principle investigator.
2. Methods
superimposed relapses (classified as primary relapsing MS) were included in the study cohort.
2.1. Study setting 2.3. Data extraction and validation Study was conducted on a hospital-based cohort. Specialized MS clinic in Hacettepe University has been active since 1995 and serves as one of the main referral centers for MS patients from all regions of Turkey. All patients were followed by one of the two MS specialists (RK or AT) since 1995. Detailed information on disease onset, demographic data, detailed neurological examination of the patient and laboratory and neuroimaging features are recorded in the baseline visit. In each follow-up visit, clinical relapses and disease course since last follow-up, current symptoms, medications, detailed neurological examination and neuroimaging findings are recorded to patient charts prospectively. In 2008, demographic data of all the previous patients who were diagnosed with MS in our center were recorded in the iMED local electronic database. In 2013, this database was updated with extensive clinical and paraclinical data and all MS patients seen in the outpatient clinic and neurology ward are recorded prospectively to the database since then. This study was approved by Hacettepe University Ethical Committee.
Data extraction workflow is as follows (Fig. 1): Patients seen between 1995 and 2015 were searched through database. 222 patients noted as “PPMS” or “PPMS-like” were detected. Of these, 30 patients were concluded as “not PPMS” (SAPMS, SPMS, RRMS, other). Next, the iMed database and all available digital and manually written patient charts were independently reviewed by two neurologists specializing on MS (AV and ED) for clinical and neuroimaging information. During this step, unclear information was clarified by phone calls to patients (n = 16). EDSS scores were derived directly from chart or determined from neurological examination. Conflicts on age of onset, type of starting symptoms and age of EDSS 6 were resolved with the primary physician of the patient (either AT or RK). In the last step, 34 patients were excluded because of inadequate information on disease onset, diagnosis or follow-up. One patient with PPMS was excluded because he had a concomitant disabling disease. Presenting symptoms were grouped under “spinal cord” referring to bilateral motor symptoms with upper motor signs, “brainstem/cerebellum” or “optic nerves” when symptoms of these regions were detected. “Supratentorial” symptoms were defined as cognitive or unilateral sensory/motor symptoms in the presence of compatible cerebral lesions on MRI and absence of bilateral motor and brainstem/cerebellar symptoms.
2.2. Patient selection Patients with a diagnosis of PPMS were identified from database. Poser or McDonald criteria were used for diagnosis depending on the year in which the patient was diagnosed. Progression of disease for at least one year from onset without any previous relapses was necessary for a patient to be classified as PPMS. Patients with a suspicious history of relapse before disease onset were excluded. Patients with
2.4. Collection of the neuroimaging data All patients were evaluated for the presence of gadolinium-enhancing lesions in brain and cervical spinal MRI in the baseline visit. 2
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
Table 1 Characteristics of patient population and comparison of patients with and without relapses and baseline gadolinium-enhancement.
Gender, F:M (n) Age of onset, y mean ± SD median, IQR Age of onset groups, n (%) 20 to <30 30 to <40 40 to <50 50< Localization of the presenting symptom, n (%) Spinal cord Brainstem/cerebellum Supratentorial Optic nerves Time to diagnosis, y mean ± SD median, IQR EDSS at diagnosis mean ± SD median, IQR Oligoclonal bandsa, n (%) Treatment status with DMT Follow-up, y mean ± SD median, IQR Age of EDSS6, y mean, 95% CI median, 95% CI Time to EDSS 6, y mean, 95% CI median, 95% CI
Total (n = 157)
PPMS with relapses (n = 35)
PPMS without relapses (n = 122)
p value
PPMS with baseline gadolinium-enhancement (n = 44)
PPMS without baseline gadolinium -enhancement (n = 113)
p value
1.5 (94/63)
1.9 (23/12)
1.4 (71/51)
0.443
1.75 (28/16)
1.4 (66/47)
0.590
34.8 ± 9.7 35.6, 27.7–41.1
32.7 ± 9.1 33, 25–37
35.4 ± 9.8 36, 27.8–42
0.139 0.104
33.7 ± 8.3 34.4, 27.7–38.2
35.2 ± 10.1 36, 26.9–42.4
0.386 0.327
122 (77.7) 44 (28) 24 (15.3) 3 (1.9)
25 (71.4) 14 (40) 3 (8.6) 2 (5.7)
97 (79.5) 30 (24.6) 21 (17.2) 1 (0.8)
0.358 0.089 0.290 0.125
32 (72.7) 20 (45.5) 5 (11.4) 3 (6.8)
90 (79.6) 24 (21.2) 19 (16.8) 0 (0)
0.395 0.005 0.467 0.021
3.1 ± 3.6 2, 1–3.9
1.7 ± 1.1 1, 1–2
3.6 ± 4 2, 1–4.9
<0.001 0.001
2.1 ± 2.2 1, 1–2
3.6 ± 4 2, 1–4.4
0.004 0.002
3.2 ± 1.3 3, 2–4 58/80 (72.5%)
2.6 ± 1.3 2.5, 2–3.5 8 (53.3) 22 (62.9)
3.4 ± 1.3 3.5, 2.5–4 50 (76.9) 88 (72.1)
0.003 0.002 0.105 0.302
2.8 ± 1.2 2.5, 2–3.5 18 (75) 27 (61.4)
3.3 ± 1.3 3.5, 2.5–4 40 (71.4) 83 (73.5)
0.029 0.017 0.792 0.174
10.2 ± 6.6 8.5, 5–14
9.8 ± 6.5 7.7, 4.8–13.9
10.4 ± 6.7 8.7, 5–14.3
0.633 0.517
8.5 ± 5 7.1, 4.2–12.5
10.9 ± 7 9.2, 5.1–15.3
0.018 0.065
42.8, 40.9–44.7 44, 40.1–47.9
42.2, 38.5–45.9 41, 37.5–44.5
43, 40.8–45.2 45, 42.3–47.7
0.633*
41.1, 37.9–44.4 40, 36.6–43.4
43.4, 41.1–45.7 45, 43.4–46.6
0.172b
9.9, 8.7–11 8, 6.7–9.3
10.7, 8.1–13.4 9, 6.7–11.2
9.5, 8.3–10.6 8, 6.6–9.4
0.459*
9.5, 7.3–11.8 7, 4.5–9.5
10, 8.7–11.3 9, 7.5–10.5
0.576b
54 (34.4) 53 (33.8) 42 (26.8) 8 (5.1)
F: female; M: male; y: years; SD: standard deviation; IQR: inter quartile range; DMT=disease-modifying therapy, CI: confidence interval. a Data for cerebrospinal fluid oligoclonal band status was available from 80 patients. b Long-rank test.
Fig. 2. Kaplan–Meier survival curves showing time to expanded disability status scale 6 (EDSS6) by age at onset (A); presence or absence of spinal symptoms at onset (B); and presence or absence of supratentorial symptoms at onset (C).
Baseline MRI is defined as the first available gadolinium-enhancing brain and spinal cord MRI of a patient. MRI scans were obtained in the study center, unless another gadolinium-enhanced MRI of adequate quality was obtained previously in another center. MRI studies of the MS patients are reported routinely by a neuroradiologist in our center and images are also reviewed during patient visit by the MS specialist in charge and recorded to the patient chart. Additionally, if a patient had MRI done in a different center the images were interpreted by one of the MS specialists and this information was also recorded in the patient's
chart and included in our study. 2.5. Statistical analysis Clinical features including gender, localization of the first symptom, oligoclonal band positivity and treatment status with disease-modifying therapies (DMTs) were compared with chi-square test. Age of onset, time to diagnosis, EDSS at diagnosis and follow-up duration were compared with independent T-test and Mann-Whitney U test. Median 3
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
Table 2 Kaplan–Meier and multivariate Cox regression analysis results of potential predictors of time to EDSS6. Factor Gender Female Male Age at onset <30 30 to <40 40 to <50 50 and older Onset symptom localization Spinal Present Absent Brainstem/ Cerebellum Present Absent Supratentorial Present Absent Clinical relapses Present Absent Gadolinium-enhancement at baseline Present Absent
Total, n
Reaching EDSS6, n (%)
Median time to EDSS6 (95% CI)
p valuea
HR (95% CI) for time to EDSS6
p value
94 63
64 (68.1) 41 (65.1)
8.0 (6.5–9.5) 9.0 (7.8–10.2)
0.587
–
ns
54 53 42 8
36 (66.7) 37 (69.8) 25 (59.5) 7 (87.5)
9.0 (7.4–10.6) 10.0 (8–12) 7.0 (5.8–8.2) 4.0 (1.5–6.5)
0.008
1.03 (1.006–1.053)b
0.012
122 35
89 (73) 16 (45.7)
8.0 (6.6–9.4) 13.0 (10.8–15.2)
0.003
2.13 (1.24–3.63)
0.006
44 113
23 (52.3) 82 (72.6)
9.0 (5.2–12.8) 8.0 (6.3–9.7)
0.085
–
ns
24 133
10 (41.7) 95 (71.4)
14.0 (12.4–15.6) 8.0 (6.8–9.3)
0.003
0.40(0.21–0.78)
0.007
35 122
23 (65.7) 82 (67.2)
9 (6.7–11.2) 8 (6.5–9.5)
0.459
–
ns
44 113
27 (61.4) 78 (69)
7 (4.5–9.5) 9 (7.5–10.5)
0.576
ns
a
Long-rank test. increased risk per year. EDSS: expanded disability status scale; HR: hazard ratio; CI: confidence interval; ns: not significant. Statistically significant p values are written in bold. b
time to EDSS6 for younger age of onset (p = 0.009) (Fig. 2A). Patients with spinal symptoms at onset had a faster progression rate (p = 0.003) (Fig. 2B). Presence of supratentorial symptoms at onset, including cognitive symptoms or sensory symptoms in the absence of bilateral motor symptoms or urinary symptoms, was associated with a longer time to EDSS6 (p = 0.002) (Fig. 2C). Gender, treatment with DMTs, presence of relapses or baseline gadolinium-enhancing lesions did not influence time to EDSS6. To study the independent effect of these factors on time to EDSS6 we introduced presence or absence of spinal and supratentorial symptoms at onset, and presence or absence of relapses and baseline gadolinium-enhancing lesions together with gender and age of onset (as continuous variable) into the Cox regression model. Older age of onset (HR value per 1-year increase=1.03, 95% CI:1.006–1.053, p = 0.012) and presence of spinal symptoms at onset (HR=2.13, 95% CI:1.24–3.63, p = 0.006) were independently associated with a shorter time to EDSS6. Presence of supratentorial symptoms at onset was independently associated with a longer time to EDSS6 (HR=0.373, 95% CI:0.186–0.745, p = 0.005) (Table 2). Presence of relapses or gadolinium-enhancing lesions on baseline MRI did not influence neither time to EDSS6 nor age at EDSS6. Regarding the age at EDSS6, younger age of onset category was associated with also a younger age at EDSS6 with both Kaplan-Meier analysis (p<0.001) and Cox regression model (HR=0.87, 95% CI: 0.84–0.89). None of the other factors examined were associated with age at EDSS6.
age of EDSS6 and median time to EDSS6 for different gender, age at onset, location of first symptom, treatment status with DMTs, baseline gadolinium-enhancing lesion status and clinical relapse status categories were calculated by Kaplan-Meier survival analysis. In the model generated for multivariate Cox regression analysis, same factors analyzed by Kaplan-Meier analysis were entered as variables, with age as continuous and others as categorical variables. To compare the mean age of onset of our population to European populations, adjusted-mean age of onset was calculated.
3. Results Study population consisted of 157 PPMS patients with a median (IQR) follow-up time of 8.5 (5.0–14.0) years. Median age of onset was 35.6 years (95% CI:27.7–41.1), median age at EDSS6 was 44 years (95% CI:40.1–47.9) and median time to EDSS6 was 8 years (95% CI:6.7–9.3). Presenting symptoms localized to spinal cord in 77.7% (n = 122), brainstem/cerebellum in 28% (n = 44), supratentorium in 15.3% (n = 24), optic nerves in 1.9% (n = 3). Among patients presenting with supratentorial symptoms, 11 had unilateral motor (7%), nine had unilateral sensory (6%) and four had combined unilateral motor and sensory symptoms in the absence of cervical lesions (3%); eight patients (5%) presented with cognitive symptoms. Other clinical characteristics of the study population is shown in Table 1. At least one clinical relapse was present in 35 cases (22.3%). Eight of these patients had two relapses during follow-up and others had one relapse. 31 (88.5%) patients had the relapse in the first five years after onset. Clinical characteristics were similar between relapsing and nonrelapsing patients, except that relapsing patients had a shorter time to diagnosis and lower EDSS at diagnosis. Gadolinium-enhancing lesions were present in the baseline MRI of 44 (28%) patients. Again, clinical characteristics were similar between patients with and without baseline gadolinium-enhancing lesions, except for a shorter time to diagnosis and a lower EDSS at diagnosis for patients with baseline gadolinium-enhancing lesions. Kaplan-Meier survival analysis revealed that time to EDSS6 was different for patients in different age of onset categories, with a longer
4. Discussion Our main findings can be summarized as: Younger age of onset is associated with a longer time to EDSS6 in PPMS. Presence of spinal symptoms at onset is associated with a faster progression rate, whereas presence of supratentorial symptoms is predictive of longer time to EDSS6 in PPMS. Having a relapsing course or gadolinium-enhancing lesion on baseline MRI does not affect disease progression rate in PPMS. Median time to EDSS6 in our study population (8 years) is within the same range with the previously reported values (7.1–14) (Table 3), 4
Multiple Sclerosis and Related Disorders 38 (2020) 101520
5.8 3
57.3 45
9.6 9
Yes Yes
No No
No Yes
No No
indicating a similar disease progression rate in our population with the others (Harding et al., 2015; Koch et al., 2009; Andersson et al., 1999; Confavreux et al., 2000; Cottrell et al., 1999; Debouverie et al., 2007; Kelly et al., 2013; Koch et al., 2015; Leray et al., 2010; Raghavan et al., 2015; Tutuncu et al., 2013). Median age of onset was younger in our cohort compared to the previous natural history studies performed in Europe and North America (38.5–44.4) (Table 3). This result can stem from the hospital-based design of our study, which may have created a referral bias towards a more active and younger-onset cases. Alternatively, Turkish population has a different age distribution compared to European and North American populations, with a higher ratio of younger individuals and lower ratio for elder, and this may have a significant effect on the average of milestone ages. Effect of relapses, gender, age of onset, and symptom of onset on time to EDSS6 in PPMS patients have been studied in several studies (Harding et al., 2015; Koch et al., 2009; Andersson et al., 1999; Confavreux et al., 2000; Cottrell et al., 1999; Debouverie et al., 2007; Koch et al., 2015; Leray et al., 2010; Stellmann et al., 2014; Confavreux et al., 2003) (Table 3). Male gender had been associated with faster progression rate in one study (Confavreux et al., 2003), however this finding was not replicated in others. Association of lower age of onset with longer time to EDSS6 in PPMS was shown in three other studies (Harding et al., 2015; Koch et al., 2009; Koch et al., 2015) and also for SPMS (Paz Soldan et al., 2015). The relationship with age of onset and rate of disease progression in the progressive phase of MS indicates that neurodegeneration at that stage is at least partly an agedependent process. However, age at onset cannot be taken as a positive prognostic factor, as it is also associated with a younger age at EDSS6 as shown in our study and previous studies (Koch et al., 2009; Koch et al., 2015). Studies on the relation of symptom of onset to pace of disease progression yielded varying results. In earlier studies, symptom type was not associated with prognosis (Cottrell et al., 1999; Debouverie et al., 2007; Leray et al., 2010; Confavreux et al., 2003). However, in two recent studies, absence of bilateral motor symptoms and presence of sensory symptoms were associated with slower disease progression to EDSS6 (Koch et al., 2009; Koch et al., 2015). In our study, onset with spinal cord symptoms which was defined as bilateral motor symptoms, was associated with shorter time to EDSS6 and onset with supratentorial symptoms was associated with longer time to EDSS6. Altogether, these findings suggest that disease onset with symptoms other than bilateral motor and brainstem/cerebellar symptoms may be associated with slower disease progression. One recent neuroimaging study reported that in PPMS patients spinal cord volume (SCV) loss progressed independent of total brain volume and T2 lesion volume change and in PPMS, and SCV was the only MRI measurement associated with EDSS increase over time, as opposed to RRMS and SPMS (Tsagkas et al., 2019). Therefore, there may be a subgroup of PPMS patients in whom the pathology is localized predominantly to supratentorium, in relation with a slower disease progression rate. This hypothesis should be further investigated by a prospective MRI study. Gadolinium-enhancing lesions were present at baseline MRI in 28% of our patients. A few other studies reported the frequency and treatment-related prognostic value of gadolinium-enhancing lesions in PPMS. In a study which was performed with triple dose gadolinium, out of 45 patients with a mean disease duration of 3.3 years, 42% had gadolinium-enhancing lesions (Ingle et al., 2005). In the PROMise trial, gadolinium-enhancing lesions were detected in 14% of all baseline MRI scans from 943 patients (Wolinsky et al., 2007). In the OLYMPUS trial, 24.5% of 439 PPMS patients had gadolinium-enhancing lesions at baseline MRI and these patients had a better response to rituximab in this study (Hawker et al., 2009). In the ORATORIO trial, 27.2% of 732 PPMS patients had gadolinium-enhancing lesions at baseline MRI. After subgroup analysis, the authors concluded that the study was underpowered to show the potential of the gadolinium-enhancing lesions as a treatment-effect–modifying factor (Montalban et al., 2017). These
2015(3) 2019
y: years; F: female; M: male; EDSS: expanded disability status scale; NS: not specified.
2.9 2 43.9 35.6 1.1 1.49 13.1 8.5 27.8 22.3 234 157
No – No – – – – – No No No No Yes No Yes No No No Yes No No No No No No No No No Yes No Yes No 10.3 8 7.1 9.9 14 10 9 – 5.6 – – – – – – 4
San Francisco, USA Ontario, Canada Lyon, France Lorraine, France British Columbia, Canada Rennes, France Calgary, Canada Hamburg & Munich, Germany Cardiff, Wales Ankara, Turkey 1999(5) 1999(7) 2000(6) 2007(8) 2009(4) 2010(11) 2015(10) 2015(14)
95 216 282 359 552 445 500 597
12.6 NS 38.6 29.8 0% NS NS NS
13.9 23.7 – 13.6 12.9 – 14 4.4
1.44 1.3 – 1.36 1.13 1.41 1.12 1.17
40.8 38.5 – 42.7 41.8 38.5 44 40.4
– 9.8 – – 7.7 – – –
– – 53 – 58.6 52.1 55 –
Relapses Symptoms Sex Age
EDSS at diagnosis (y) Time to diagnosis (y) Age of onset (y) F:M Follow-up (y) PRMS (%) n Location Year (Reference)
Table 3 Previous studies providing data on predictors of progression for primary progressive multiple sclerosis.
Age of EDSS=6 (y)
Time to EDSS=6 (y)
Prognostic factors (time to EDSS6)
A. Vural, et al.
5
Multiple Sclerosis and Related Disorders 38 (2020) 101520
A. Vural, et al.
results show that, gadolinium-enhancing lesions are not rare in PPMS and they may have a prognostic value regarding treatment response, however this topic requires further investigation. There are several limitations of our study: One of the main limitations is the retrospective nature of the study, which is inevitable for such long-term studies. Also, although our study population was large enough to show the effect of age of onset and onset symptoms on progression rate, we cannot rule out that larger studies may show a relationship between baseline gadolinium-enhancement and prognosis. In conclusion, our study confirms the association between the age of onset and symptom at onset with time to EDSS6. Population-based natural history studies are necessary to investigate further the characteristics of PPMS in different populations.
Multiple sclerosis with a progressive course from onset in Lorraine-eastern France. J. Neurol. 254 (10), 1370–1375. Harding, K.E., Wardle, M., Moore, P., Tomassini, V., Pickersgill, T., Ben-Shlomo, Y., et al., 2015. Modelling the natural history of primary progressive multiple sclerosis. J. Neurol. Neurosurg. Psychiatry 86 (1), 13–19. Hawker, K., O'Connor, P., Freedman, M.S., Calabresi, P.A., Antel, J., Simon, J., et al., 2009. Rituximab in patients with primary progressive multiple sclerosis: results of a randomized double-blind placebo-controlled multicenter trial. Ann. Neurol 66 (4), 460–471. Ingle, G.T., Sastre-Garriga, J., Miller, D.H., Thompson, A.J, 2005. Is inflammation important in early PPMS? a longitudinal MRI study. J. Neurol. Neurosurg. Psychiatry 76 (9), 1255–1258. Kelly, S.B., Kinsella, K., Duggan, M., Tubridy, N., McGuigan, C., Hutchinson, M, 2013. A proposed modification to the Mcdonald 2010 criteria for the diagnosis of primary progressive multiple sclerosis. Mult. Scler. 19 (8), 1095–1100. Koch, M., Kingwell, E., Rieckmann, P., Tremlett, H, 2009. The natural history of primary progressive multiple sclerosis. Neurology 73 (23), 1996–2002. Koch, M.W., Greenfield, J., Javizian, O., Deighton, S., Wall, W., Metz, L.M, 2015. The natural history of early versus late disability accumulation in primary progressive MS. J. Neurol. Neurosurg. Psychiatry 86 (6), 615–621. Leray, E., Yaouanq, J., Le Page, E., Coustans, M., Laplaud, D., Oger, J., et al., 2010. Evidence for a two-stage disability progression in multiple sclerosis. Brain 133 (Pt 7), 1900–1913. Montalban, X., Hauser, S.L., Kappos, L., Arnold, D.L., Bar-Or, A., Comi, G., et al., 2017. Ocrelizumab versus placebo in primary progressive multiple sclerosis. N. Engl. J. Med. 376 (3), 209–220. Paz Soldan, M.M., Novotna, M., Abou Zeid, N., Kale, N., Tutuncu, M., Crusan, D.J., et al., 2015. Relapses and disability accumulation in progressive multiple sclerosis. Neurology 84 (1), 81–88. Raghavan, K., Healy, B.C., Carruthers, R.L., Chitnis, T, 2015. Progression rates and sample size estimates for PPMS based on the climb study population. Mult. Scler. 21 (2), 180–188. Sorensen, P.S., Blinkenberg, M, 2016. The potential role for ocrelizumab in the treatment of multiple sclerosis: current evidence and future prospects. Ther. Adv. Neurol. Disord. 9 (1), 44–52. Stellmann, J.P., Neuhaus, A., Lederer, C., Daumer, M., Heesen, C, 2014. Validating predictors of disease progression in a large cohort of primary-progressive multiple sclerosis based on a systematic literature review. PLoS ONE 9 (3), e92761. Tsagkas, C., Magon, S., Gaetano, L., Pezold, S., Naegelin, Y., Amann, M., et al., 2019. Preferential spinal cord volume loss in primary progressive multiple sclerosis. Mult. Scler. 25 (7), 947–957. Tutuncu, M., Tang, J., Zeid, N.A., Kale, N., Crusan, D.J., Atkinson, E.J., et al., 2013. Onset of progressive phase is an age-dependent clinical milestone in multiple sclerosis. Mult. Scler. 19 (2), 188–198. Wolinsky, J.S., Narayana, P.A., O'Connor, P., Coyle, P.K., Ford, C., Johnson, K., et al., 2007. Glatiramer acetate in primary progressive multiple sclerosis: results of a multinational, multicenter, double-blind, placebo-controlled trial. Ann. Neurol. 61 (1), 14–24.
Declaration of Competing Interest Atay Vural, Eda Derle, Güliz Sayat-Gürel, Rana Karabudak and Aslı Tuncer does not have any conflict of interest to report related to the study. Acknowledgements We are thankful to our patients and their care-givers for participation in the study. This study is dedicated to the memory of our beloved colleague and dear friend Güliz Sayat-Gürel. References Andersson, P.B., Waubant, E., Gee, L., Goodkin, D.E, 1999. Multiple sclerosis that is progressive from the time of onset: clinical characteristics and progression of disability. Arch. Neurol. 56 (9), 1138–1142. Confavreux, C., Vukusic, S., Adeleine, P, 2003. Early clinical predictors and progression of irreversible disability in multiple sclerosis: an amnesic process. Brain 126 (Pt 4), 770–782. Confavreux, C., Vukusic, S., Moreau, T., Adeleine, P, 2000. Relapses and progression of disability in multiple sclerosis. N. Engl. J. Med. 343 (20), 1430–1438. Cottrell, D.A., Kremenchutzky, M., Rice, G.P., Koopman, W.J., Hader, W., Baskerville, J., et al., 1999. The natural history of multiple sclerosis: a geographically based study. 5. the clinical features and natural history of primary progressive multiple sclerosis. Brain 122 (Pt 4), 625–639. Debouverie, M., Louis, S., Pittion-Vouyovitch, S., Roederer, T., Vespignani, H, 2007.
6