Higher Incidence and Risk of Developing Second Solid Cancers (SSC) after Haematopoetic Stem Cell Transplantation (HSCT) As Compared to General Population

Introduction: As the number of long-term survivors following HSCT is increasing, the long-term risks and associated morbidity has become important component of survivorship program. The known risk factors for developing cancer include use of chemotherapy agents, radiation exposure, immune dysfunction, previous malignancy in addition to other factors and as HSCT process involves all these factors, this single centre retrospective analysis was undertaken to evaluate the risk of developing SSC in the patients receiving transplant.

Methods: From February 1982 to February 2016, 2231 patients received 2495 transplants (median age: 46yr., range: 14-76 yr.; M: 1586, F: 909) for haematological malignancies (Leuk: 744, lymphoma:767, myeloma:848, solid tumours/other:136). Donor was allogeneic (n=744) or autologous (n=1751) and conditioning was with (n=614) or without TBI (n=1881). Donor was sibling (n=375), matched unrelated (n=355), haploidentical relative (n=3) or umbilical cord blood (n=11). Source of stem cell was marrow (n=367), PBSC (n=2086), both (n=31) or cord blood (n=11). GVH prophylaxis included Campath or ATG in 369 cases. Of all the patients 1985 received single transplant, 231 had two, 13 had 3 and 2 had 4 HSCT procedures. Data was analysed as of 15/04/2016 using competing risk model with death as the competing event. Comparison of incidence to general population was performed by computing standardized incidence rates (SIR). Patients with second haematological malignancy were not included in this analysis.

Results: Median follow-up was 5.3 years (range: 0-32 years). Patient follow-up was more than 10 years in 467 cases (19%), between 5 to 10 years in 430 (17%), 2 to 5 years in 607 (24%) and less than 1 year in 997 cases (40%). 36% patients were followed-up for more than 5 years. Second solid cancers developed in 116 patients with the incidence of 1% at 5yr (95% CI: 0.5-2.6), 3% at 10 yr (95% CI: 1.6-5.3), 6% at 15yr (95%CI: 3.6-8.8) and 10% (95% CI: 5.9-15.5) at 20 years. Median time to develop SSC from date of HSCT was 11 yr (range: 0.4-28.1 yr). Primary site for SSC included skin (n=37), breast (n=22), GI (n=15), GU (n=16), H&N (n=10), lung (n=6), CNS (n=4), Endocrine (n=4) & HPB (n=2). There was no difference with type of transplant i.e. auto or allograft. Autograft and allograft groups were analysed separately. In univariate analysis, allograft group showed higher cumulative incidence of SSC with use of PBSC (p<0.0001), campath/ATG (p=0.0002), donor other than sibling (P=0.0004), RIC (p<0.0001), non-TBI conditioning (p=0.007), older age at transplant (0.008), development of agvhd or cgvhd (p=0.023) and transplant year after 2000 (p=0.01). In multivariate analysis age above 50 (RR: 1.8, 95%CI: 1.2-1.8, p=0.046) and RIC (RR: 4.4, 95% CI: 1.2-8.3, p=0.03) were independently associated with higher risk of SSC. In autograft group, there were no independent risk factors in univariate or multivariate analysis. As compared to general population incidence was higher for all cancers (SIR=7.4) and also cancers at every primary site. Risk was highest for breast (SIR=14.3), Head/neck (SIR=25.6), brain (SIR=17.4) and colon (SIR=6.2). Overall survival is significantly shorter in patients who develop SSC (median: 12 yr vs. not reached, p<0.0001). The median time to develop SSC from the date of HSCT has significantly shortened over last three decades (years 1982-1990: 21yr, 1991-2000: 11.1 yr, 2001-2010: 6.34 yr, 2011-2016: 2.2 yr; p=0.0001)

Conclusion: This single centre analysis confirms that the risk of developing SSC increases with advancing age, use of RIC allograft, longer follow-up and leads to inferior survival. Since the year 2000, SSC are developing early after transplant and it needs to be evaluated if this is a trend seen at other centers and if so, is it related to increasing use of RIC, increasing number of elderly patients, severity of immune-suppression or higher incidence of GVHD.