Pathol. Oncol. Res. (2018) 24:469–475 DOI 10.1007/s12253-017-0266-7 ORIGINAL ARTICLE Long-Term Outcome of Autologous Hematopoietic Stem Cell Transplantation (AHSCT) for Acute Myeloid Leukemia (AML)- Single Center Retrospective Analysis 1 1 1 1 Grzegorz Helbig & Anna Koclęga & Krzysztof Woźniczka & Małgorzata Kopera & Sławomira Kyrcz-Krzemień Received: 4 October 2016 /Accepted: 21 June 2017 /Published online: 28 June 2017 The Author(s) 2017. This article is an open access publication Abstract For patients with acute myeloid leukemia (AML) in with acceptable toxicity profile. Leukemia recurrence complete remission without an acceptable HLA donor, the remained the leading cause of death. autologous hematopoietic stem cell transplantation (AHSCT) may remain a therapeutic option as remission con- Keywords Acute myeloid leukemia Autologous . . solidation, however its role is still a subject of continued de- hematopoietic stem cell transplantation Relapse Outcome bate. One hundred and twenty patients who underwent AHSCT for AML were included in this retrospective single center analysis. The procedure was performed over a 19 years Introduction period and transplanted patients were in first complete remis- sion (CR1; n = 109) or in second CR (CR2; n = 11). The An intensive combined chemotherapy including anthracycline median age at transplant was 37 years (range 18–64). The and cytarabine remains a mainstay of induction treatment for source of stem cells was bone marrow (n = 61; 50.8%), pe- medically fit and younger (<60 years) patients with acute my- ripheral blood (n = 36; 30%) and bone marrow with peripheral eloid leukemia (AML) producing a complete remission (CR) blood (n = 23; 19.2%). The median time from AML diagnosis rate ranges from 60% to 70% . The continuation of post- to AHSCT was 0.8 year (range 0.3–4.4) and the median remission therapy seems to be mandatory in order to prevent follow-up after AHSCT for surviving patients was 12.8 years relapse which may occur in virtually all patients after CR (range 3.1–20.5). The median LFS was 1.1 year. The proba- achievement . The optimal post-remission treatment is bility of LFS calculated at 5 years and 10 years after trans- still a matter of debate and may include repeated cycles of plantation was 28% (95%CI, 22%–32%) and 21% (95%CI, high-dose cytarabine, autologous hematopoietic stem cell 18%–24%), respectively. The last relapse occurred 14.8 years transplantation (AHSCT) and allogeneic hematopoietic after AHSCT and among patients who survived >2 years, stem cell transplantation (AlloHSCT). The choice of 28.4% (27/95) had leukemia recurrence. The median OS post-remission therapeutic strategy should be guided by was 1.7 years. The probability of OS after 5 years and 10 years several factors associated with disease (e.g. cytogenetics), was 29% and 22%, respectively. There was a tendency for patient (e.g. performance status) or transplant (e.g. donor increased LFS for patients younger than 50 years at transplant availability). The strongest anti-leukemic effect is associ- if compared to older population. AHSCT for AML was safe ated with AlloHSCT, however it is counterbalanced by high non-relapse treatment-related mortality . The role of AHSCT as a post-remission therapy for AML patients remains unclear. The toxicity of AHSCT seems to be comparable with that of intensive chemotherapy. The main * Grzegorz Helbig firstname.lastname@example.org concern is still associated with a high leukemia recurrence rate which usually occurs within the first 2 years after transplant 1 with good prognosis for those who survived after this period School of Medicine in Katowice, Department of Hematology and . Nevertheless, the cumulative relapse incidence at 10 years Bone Marrow Transplantation, Medical University of Silesia, Dąbrowski street 25, 40-032 Katowice, Poland was 16% in the recent report of the European Society for 470 Helbig G. et al. Blood and Marrow Transplantation (EBMT). Older patient Statistical Methods age, peripheral blood as a source of stem cells and M0, M6 and M7 leukemia subtypes were found to be associated with The probability OS and LFS were calculated according to higher relapse rate in the multivariate analysis . Kaplan-Meier estimate. The Kruskal-Wallis test was used to Overall, the patients autografted in CR1 had superior compare more than two independent groups of variables. The leukemia-free survival (LFS) and lower leukemia recurrence, probability of LFS was defined as the time from transplant to but not overall survival (OS) if compared with those treated with relapse or death in CR. All calculations were made from the intensive chemotherapy. The summary of numerous studies has date of transplantation. Comparisons between the variables demonstrated that AHSCT should be offered for younger pa- were carried out by log-rank test. Statistical significance was tients in the favorable and intermediate cytogenetic risk groups, defined at a P value <0.05. Transplant-related mortality however its optimal role requires further investigations . (TRM) was defined as death within 100 days after AHSCT Herein we report the outcome of adults patients with not related to disease or relapse. Data analysis was censored at AML who underwent AHSCT in our institution over a the time of AlloHSCT in 7 patients who relapsed. 19-year period. Results Material and Methods The median time from AML diagnosis to AHSCT was Patients and Procedures 0.8 year (range 0.3–4.4) and the median follow-up after AHSCT for surviving patients was 12.8 years (range 3.1– One hundred and twenty patients who underwent AHSCT for 20.5). Twenty three patients were alive at the time of anal- AMLwereincludedinthisretrospective single center analysis. ysis. The median number of transplanted CD34-positive The procedure was performed between years 1990 and 2009 and cells was 2.02 (range 0.39–13.3). Absolute neutrophil transplanted patients were in first complete remission (CR1) or in count (ANC) >0.5 G/L and platelet (PLT) count >50 G/L second CR (CR2). The median age at transplant was 37 years were achieved after median of 21 days (range 11–120) and (range 18–64) and 13.3% of patients were >50 years at the time 45 days (11–390), respectively. The fastest ANC recovery of procedure. There was a slight male predominance (51.6%). was observed in patients transplanted from PB (med. 56% of AML patients were diagnosed before the year 2000. The 17 days; range 11–37), if compared with PB/BM (med. diagnosis of AML was established according to the French- 23 days; range 11–42) and BM (med. 29 days; range 11– American-British (FAB) classification. The distribution between 120) p < 0.001. In regard to PLT count, the fastest recovery studied patients was as follows: M0 in 1.6% (n =2), M1 in was demonstrated for PB (med. 18 days; range 11–160), if 19.1% (n = 23), M2 in 34.1% (n = 41), M4 in 34.1% (n =41), compared with BM (med. 55 days; range 11–390) and PB/ M5 in 8.3% (n=10),M6in1.6%(n = 2) and M7 in 0.8% (n=1). BM (med. 79 days, range 20–180) p =0.003. The patients with M3 were excluded from analysis. The Five patients died within the first 100 days after transplan- cytogenetic/molecular results were available for merely 27 pa- tation; all deaths were due to severe infectious complications. tients (22.5%) and 8/27 (29.6%) subjects were assigned to favor- More than half of transplanted patients suffered from moder- able [6 pts. with t(8;21) and 2 pts. with inv.(16)] and 14/27 ate stomatitis and pharyngitis (n = 66). 23 patients had fever of (51.8%) to intermediate (all with normal diploid karyotype with- unknown origin. Bacteremia was present in 15 subjects and out FLT3-ITD mutation) and 5/27 (18.5%) to adverse risk cate- 33% of them developed severe bilateral pneumonia leading to gory(3pts.withdetectableFLT3-ITDmutationand2pts.with death. The other complications included herpes zoster reacti- monosomy 7). Prior myelodysplastic syndrome was diagnosed vation (n = 7), proctitis (n = 6), hepatitis (n = 6) and urinary in 9.1% of patients. The vast majority of patients were tract infections (n =3). transplanted in CR1 (90.8%; n = 109) whereas the remaining The median LFS was 1.1 year. The probability of LFS 11 subjects (9.2%) had CR2. The source of stem cells was bone calculated at 5 years and 10 years after transplantation was marrow (n = 61; 50.8%), peripheral blood (n = 36; 30%) and 28% (95% confidence interval [95%CI], 22%–32%) and bone marrow with peripheral blood (n = 23; 19.2%). Graft 21% (95%CI, 18%–24%), respectively. The last relapse oc- ex vivo purging was not performed. The assessment of minimal curred 14.8 years after AHSCT and among patients who sur- residual disease (MRD) in collected product using flow cytome- vived >2 years, 28.4% (27/95) had leukemia recurrence. try or molecular tests was not a routine practice at that time in our The median OS was 1.7 years. The probability of OS after center. Most transplants were performed before the year 2000 5 years and 10 years was 29% (95%CI, 25%–33%) and 22% (52.5%; n = 63) and the commonest conditioning regimen (95%CI, 19%–25%), respectively (Figs. 1 and 2). No factor consisted of busulfan and cyclophosphamide (84.1%; n = 101). was found to have a significant impact on LFS and OS in Details were shown in Table 1. univariate analysis (Table 2). There was a tendency for better Long-term outcome of autologous hematopoietic stem cell transplant 471 Table 1 Study patients Patients characteristics Number of patients: 120 characteristics Age at transplant (y; median; range) 37 (18–64) >60 y 3 (2.5%) Between 50 and 59 y 13 (10.8%) <50 y 104 (86.6%) Gender female/male 58/62 Date of AML diagnosis >2000 y 53 (44.1%) Prior MDS 11 (9.1%) FAB classification M0 2 (1.6%) M1 23 (19.1%) M2 41 (34.1%) M4 41 (34.1%) M5 10 (8.3%) M6 2 (1.6%) M7 1 (0.8%) AML status at transplant CR1 109 (90.8%) CR2 11 (9.2%) Year of transplant >2000 y 57 (47.5%) Source of stem cells Bone marrow 61 (50.8%) Peripheral blood 36 (30.0%) Bone marrow and peripheral blood 23 (19.2%) Type of conditioning Chemotherapy-based 120 (100%) Busulfan/cyclophosphamide 101 (84.1%) Busulfan and cytarabine 4 (3.3%) Etoposide/cytarabine/cyclophosphamide 15 (12.5%) Number of transplanted CD34-positive cells (median; range) 2.02 (0.39–13.3) Time to ANC recovery >0.5 G/L (days; range) 21 (11–120) Time to PLT recovery >50 G/L (days; range) 45 (11–390) Number of RBCs transfusions (units; median; range) 3 (0–33) Number of PLT transfusions (units; median, range) 5 (0–42) Death to day +100 after AHSCT 5 (4.1%) AML acute myeloid leukemia, ANC absolute neutrophil count, AHSCT autologous hematopoietic stem cell transplantation; BM bone marrow, CR complete remission, FAB French-American-British, MDS myelodysplastic syndrome, PB peripheral blood; PLT platelets, RBC red blood cells LFS in patients younger than 50 years at transplant if com- are alive in CR whereas the remaining 4 died due to post- pared with older population (32% [95%CI 28–36] vs 6% transplant complications (Table 3). [95%CI 0–12]; p =0.06). Cytogenetic and molecular results were available only for 27 patients and probably for that reason, no difference be- Discussion tween risk categories in terms of LFS rates was demonstrated. The probability of LFS calculated at 5 years after AHSCT for For patients with AML without an acceptable HLA donor, the favorable, intermediate and adverse risk groups was follow- AHSCT may remain an option for post-remission consolida- ing: 37%, 30% and 20%, respectively (p =0.9). tion. In the present study, we analyzed the results of our single Seven patients who relapsed after AHSCT were proceeded institution experience in AHSCT for AML patients. A total of to AlloHSCT from matched unrelated donors. Three of them 120 patients with a median age of 37 years were autografted 472 Helbig G. et al. OS for AML after AH SC T Fig. 1 Overall survival curve for 1,0 AML patients after autologous transplantation 0,9 0,8 0,7 0,6 0,5 0,4 Median s ur v iv al: 1.6y OS at 5y = 29%; at 10y = 22% 0,3 0,2 0,1 0,0 0 5 10 15 20 25 Time in y ear s over a 19 year period. The vast majority of patients were Historically, the use of bone marrow (BM) as a source of transplanted in CR1 (91%) and 23 survivors were followed stem cells was associated with difficulties in stem cell collec- for an average of 13 years with a maximum follow-up of more tion and delayed hematopoietic recovery . The introduction than 20 years. The median LFS and OS survivals were of peripheral blood (PB) progenitor cells as a source of stem 13 months and 19 months, respectively and they were com- cells allowed for better collection yield, faster engraftment and parable with those reported by others . It should be stressed reduction of infection risk . 70% of patients (n = 84) from that the last relapse occurred almost 15 years after AHSCTand our study received BM-derived stem cells, however almost 27 (28%) patients had late (>2 years) leukemia recurrence. 30% of them (n = 23) yielded an insufficient number of cells LFS for AML after AHSCT Fig. 2 Leukemia-free survival curve for AML patients after 1,0 autologous transplantation 0,9 0,8 0,7 0,6 0,5 0,4 Median LFS: 1.1 y 0,3 LFS at 5y=28%; at 10y=21% 0,2 0,1 0,0 0 5 10 15 20 25 Time in years Leukemia-free survival Overall survival Long-term outcome of autologous hematopoietic stem cell transplant 473 Table 2 Prognostic factors in univariate analysis 5 years. Moreover, these authors found age and karyotype as predictors of poor outcome. Adverse cytogenetic risk also Factor 5y OS (95%CI) 5y LFS (95%CI) correlated with a decreased probability of LFS and OS in a Age large EBMT study published recently . In contrary, no cy- ≥ 50y 12 (4–20) 6 (0–12) togenetic abnormality nor any other tested factors were iden- < 50y 32 (28–36) 32 (28–36) tified as the predictors for LFS and OS in our study. However, p =0.19 p =0.06 it should be mentioned that majority of our patients had un- available cytogenetic data. The reason for that can partially be Gender Female 27 (22–32) 28 (22–34) related to the fact that about 60% of patients were diagnosed before the year 2000 when the tests were not routinely per- Male 31 (26–36) 28 (23–33) formed in our country. p =0.9 p =0.88 It has been demonstrated that patients who were leukemia- Date of AML diagnosis free more than 2 years after transplant less likely had disease < 2000y 30 (25–35) 28 (23–32) recurrence [4, 7]. However, this statement is becoming false in ≥ 2000y 28 (22–34) 25 (19–31) the light of the recently published paper by Czerw et al. . p =0.89 p =0.93 The recurrence incidence was 16% at 10 years and this finding Prior MDS was in line with that presented by other groups . To con- Yes 22 (9–35) 22 (9–35) clude, disease relapse seems to be the leading cause of death in No 30 (26–34) 29 (25–33) autografted AML patients. The same is also true for our study p =0.62 p =0.68 cohort. All fatal outcomes were due to leukemia recurrence FAB classification with subsequent resistance to chemotherapy. The long-term M1 30 (20–40) 30 (20–40) survival was seen only in patients who underwent M2 36 (29–43) 36 (29–43) AlloHSCT, but it also was not a rule. It was reported by others M4 24 (17–31) 23 (17–29) that higher probability of relapse was associated with PB as a M5 20 (8–32) 12 (1–21) source of stem cells, probably due to graft contamination with p =0.36 p =0.74 leukemic cells . No difference between PB and BM in AML status at transplant terms of disease relapse was found in our study. CR1 30 (26–34) 29 (25–33) Despite several published studies a definitive place for CR2 11 (1–21) 11 (1–21) AHSCT in AML patients in CR1 is a subject of continued p =0.23 p =0.33 debate. A large prospective randomized trial of remission Year of transplant consolidation with AHSCT versus chemotherapy for ≥ 2000y 26 (20–32) 26 (20–32) AML has been published recently. 517 patients were in- < 2000y 31 (26–36) 30 (25–35) cluded over a 11-year period. This study has demonstrated p =0.93 p =0.77 reduced relapse rate (58% vs 70%) and tendency to better Source of stem cells LFS (38% vs 29%) in AHSCT arm. OS were comparable BM 33 (27–39) 31 (25–37) (44% vs 41%) between arms at 5 years from randomiza- PB 24 (17–31) 22 (15–29) tion. 17% of patients in AHSCT arm if compared with BM + PB 27 (18–36) 25 (16–34) 25% in chemotherapy arm required a rescue procedure P =0.92 p =0.24 with AlloHSCT . A continued sub-analysis of AML patients aged 40–60 years showed comparable OS follow- AML acute myeloid leukemia, BM bone marrow, CR complete remission, ing AHSCT and AlloHSCT in the intermediate cytogenet- FAB French-American-British, MDS myelodysplastic syndrome, PB pe- ripheral blood ic risk group, however LFS was significantly increased in the latter . Treatment-related mortality after AHSCT was 4% . The above-mentioned results compare well and required an extra growth factor -stimulated PB collection. with those presented by Keating et al. The authors com- In fact, the fastest ANC and PLT recovery was demonstrated pared LFS and OS rates for patients receiving AlloHSCT for patients transplanted from PB if compared with those re- vs AHSCT for AML in CR1 and they did not find any ceiving stem cells form BM and BM/PB (p < 0.001 and difference between studied groups. A cumulative inci- p < 0.003, respectively). The infection rate did not differ be- dence of relapse at 5 years was significantly higher for tween groups (data not shown). The LFS and OS curves were AHSCT than AlloHSCT whereas a significantly lower also comparable. treatment-related mortality rate at 5 years was demonstrat- If we compare our results with those reported by Collison ed for the former group . The German AML96 Study has found a significantly better OS in autografted patients et al. , we found a comparable probability of LFS and OS at 474 Helbig G. et al. Table 3 Patient undergoing Patient ID Gender AML Age at Time between AHSCT Time from Current AlloHSCT after relapse post FAB AHSCT (y) and AlloHSCT (y) ALLOHSCT (y) status AHSCT KB F 2 22 14 1.7 Alive in CR TC M 5 31 0.3 11.4 Alive in CR KK F 4 19 0.7 NA Dead LM M 4 37 2.5 NA Dead RS M 1 30 0.5 NA Dead RW F 1 43 1.6 5.2 Alive in CR EP F 2 48 0.4 NA Dead AML acute myeloid leukemia, AHSCT autologous hematopoietic stem cell transplantation, AlloHSCT allogeneic hematopoietic stem cell transplantation, CR complete remission, FAB French-American-British, F female, M male, NA not applicable in the intermediate cytogenetic risk group if compared assigned to intermediate/adverse risk groups in our center with chemotherapy and AlloHSCT arms (62% vs 41% is about 60% (data not shown). vs 44%, respectively) . Interestingly, the results of In summary, the autografting for AML was safe with ac- cytogenetic studies did not influence the post-transplant ceptable post-transplant toxicity. If compared with chemother- outcome in the Keating study  and the reason for that apy, AHSCT reduced relapse risk and gives better LFS with is difficult to explain. comparable OS. The major concern of AHSCT in AML pa- We present long-term data of AML patients who were tients results from high rate of leukemia recurrences. It seems autografted in CR1/CR2. All those patients did not have a reasonable to elaborate the strategies which might prevent matched related- or unrelated donor at the time of deci- disease relapse. The introduction of minimal residual disease sion. One should keep in mind that more than 50% of assessment may allow for early introduction of preventive patients were diagnosed and transplanted before the year strategies . AlloHSCT should be offered for relapsed pa- 2000 when the availability of volunteer donors was limit- tients who achieved subsequent CR. ed, especially in Eastern and Central Europe. The major Compliance with Ethical Standards concern of our study was associated with missing cytoge- netic data. As a consequence, unselected AML population Conflict of Interest The authors declare that they have no conflict was proceeded to transplant and decision was not directed of interest. by genetic risk. Surely, it had a great impact on data in- terpretation and transplant outcome. Moreover, MRD as- Ethical Approval For this type of study formal consent is not required. sessment in collected product was not a routine practice Open Access This article is distributed under the terms of the Creative and the infusion of MRD-positive cells was likely. A pro- Commons Attribution 4.0 International License (http:// portion of our patients was in CR2 at transplant. creativecommons.org/licenses/by/4.0/), which permits unrestricted use, Interestingly, the LFS and OS were comparable between distribution, and reproduction in any medium, provided you give appro- subjects transplanted in CR1 and CR2. Of note is, that the priate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. number of patients in CR2 was small (<10%) to draw conclusions. A vast majority of our patients were less than 60 years at transplant. The results of AHSCT for older References AML population are scarce. It was demonstrated that such patients had the 3-year LFS and OS of 28% and 39%, respectively . 1. Dohner H, Estey EH, Amadori S et al (2010) Diagnosis and man- agement of acute myeloid leukemia in adults: recommendations One should be aware that the current policy of our from an international expert panel, on behalf of the European center is to perform AHSCT only in AML patients being LeukemiaNet. Blood 115:453–474 assigned to favorable cytogenetic risk group with MRD 2. Cassileth PA, Harrington DP, Hines JD et al (1988) Maintenance negativity of transplanted product. The patients with in- chemotherapy prolongs remission duration in adult acute nonlymphocytic leukemia. J Clin Oncol 6:583–587 termediate and adverse cytogenetic/molecular features are 3. Schlenk RF (2014) Post-remission therapy for acute myeloid leu- proceeded to AlloHSCT. Of note is, that in our center kemia. Haematologica 99:1663–1670 there are no patients left without transplantation after 4. Breems DA, Lowenberg B (2007) Acute myeloid leukemia and the completion of post-remission consolidation. The probabil- position of autologous stem cell transplantation. Semin Hematol 44: ity of OS at 2 year after AlloHSCT for AML patients 259–266 Long-term outcome of autologous hematopoietic stem cell transplant 475 4. Breems DA, Lowenberg B (2007) Acute myeloid leukemia and the 10. Bhatia S, Robison LL, Francisco L et al (2005) Late mortality in survivors of autologous hematopoietic cell transplantation: report position of autologous stem cell transplantation. Semin Hematol 44: 259–266 from the bone marrow transplant survivor study. Blood 105:4215– 5. Czerw T, Labopin M, Gorin NC et al (2016) Long-term follow-up 11. Vellenga E, van Putten W, Ossenkoppele GJ et al (2011) of patients with acute myeloid leukemia surviving and free of dis- Autologous peripheral blood stem cell transplantation for acute ease recurrence for at least 2 years after autologous stem cell trans- myeloid leukemia. Blood 118:6037–6042 plantation: a report from the acute leukemia working Party of the 12. Cornelissen JJ, Versluis J, Passweg JR et al (2015) Comparative European Society for blood and marrow Transplnatation. Cancer therapeutic value of post-remission approaches in patients with acute 122:1880–1887 myeloid leukemia aged 40-60 years. Leukemia 29:1041–1050 6. Zuckerman T, Beyar-Katz O, Rowe JM (2016) Should 13. Keating A, DaSilva G, Perez WS et al (2013) Autologous autotranplantation in acute myeloid leukemia in first complete re- blood cell transplantation versus HLA-identical sibling trans- mission be revisited. Curr Opin Hematol 23:88–94 plantation for acute myeloid leukemia in first complete re- 7. Collisson EA, Lashkari A, Malone R, Paquette R, Emmanouilides mission: a registry study from the Center for International C, Territo MC, Schiller GJ (2003) Long-term outcome of autolo- Blood and Marrow Transplantation Research. Haematologica gous transplantation of peripheral blood progenitor cells as 98:185–192 postremission management of adult myelogenous leukemia in first 14. Phirrmann M, Ehninger G, Thiede C et al (2012) Prediction of complete remission. Leukemia 17:2183–2188 postremission survival in acute myeloid leukemia: a posthoc anal- 8. Mehta J, Powles R, Singhal S et al (2000) Autologous bone marrow ysis of the AML96 trial. Lancet Oncol 13:207–214 transplantation for acute myeloid leukemia in first remission: iden- 15. Thomas X, Suciu S, Rio B et al (2007) Autologous stem cell trans- tification of modifiable prognostic factors. Bone Marrow plantation after complete remission and first consolidation in acute Transplant 1995; 16: 499–506 myeloid leukemia patients aged 61-70 years: results of the prospec- 9. Bensinger WI, Deeg HJ (2000) Blood or marrow? Lancet 355: tive EORTC-GIMEMA AML-13 study. Haematologica 92:389– 1199–1200
Pathology & Oncology Research – Springer Journals
Published: Jun 28, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera