Subsite specific risk of colorectal cancer risk in patients treated for hodgkin's lymphoma: A long-term follow-up study in the Netherlands

Michael Schaapveld, Cecile P. Janus, Jan Paul De Boer, Augustinus Krol, Josée M. Zijlstra, John Raemaekers, Leontien C. Kremer, Marieke W. Louwman, Otto Visser, Marie L. De Bruin, Berthe M. Aleman, Flora E. Van Leeuwen, Monique E. Van Leerdam

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Background and objective After the introduction of multi-agent chemotherapy (CT) and modern radiotherapy (RT) in the last decades, Hodgkin's lymphoma (HL) has become the prototype of a curable malignancy. As a result there is a growing cohort of long-term HL survivors at risk for treatment-related long-term complications, including second malignancies. While many studies reported increased risks of breast and lung cancer after HL, few studies examined the risk of colorectal cancer (CRC). We assessed the long-term risk of CRC after HL treatment, focusing on the risks for colon subsites and rectum, taking into account RT fields and specific CT agents. Methods Our multicenter Dutch cohort comprises 2,820 5-year HL survivors, diagnosed before age 51 and treated between 1965 and 1995. Incidence of CRC was compared with general population rates by calculating standardized incidence ratios (SIRs) and absolute excess risks (AERs). Treatment effects were quantified in competing risk regression analyses. Results After a median follow-up of 21.5 years, 49 patients were diagnosed with CRC. The SIR for CRC was 2.7 (95% confidence interval (CI) 2.0-3.6), with 6.3 excess cases per 10,000 patient-years. The cumulative incidence of CRC, accounting for death as a competing risk, was 0.8% at 20 years and 2.2% at 30 years of follow-up. The highest SIR was observed for cancer of the transverse colon (SIR 7.5 95% CI 3.7-13.4). The SIR of CRC increased strongly with decreasing age at first treatment (SIR 4.5 (95% CI 2.1-7.6) and SIR 2.3 (95% CI 1.4-3.4) for patients aged , 25 years and ≥ 35 years, respectively). Especially for colon cancer, the SIR increased with longer follow-up duration (in 30-year survivors; SIR 4.1, 95% CI 2.0-7.5, AER 25.9). The highest risk of CRC (SIR 5.5, 95% CI 3.5-8.4) was observed in patients who were treated with infradiaphragmatic RT and CT. Especially patients treated with para-aortic and spleen fields had a strongly increased risk of cancer of the transverse colon (SIR 17.1, 95% CI 4.7-43.8). The SIR of rectal cancer was most elevated following RT to para-aortic and iliac lymph nodes (SIR 6.3, 95% CI 2.3-13.7). In competing risk regression analyses patients treated with high-dose procarbazine (.4.2g/m2) had a significantly increased risk of CRC (subdistribution Hazard ratio (sHR) 2.4 (95%CI 1.3-4.7)) compared with patients treated without procarbazinecontaining CT. The sHR of CRC was 3.9 (95%CI 1.5-9.8) for patients that received procarbazine- containing CT and infradiaphragmatic RT compared with patients that received none of these treatments. Conclusions Infradiaphragmatic RT and high dose procarbazine are associated with a strongly increased risk of CRC in HL patients. CRC surveillance can be considered in this high-risk group.
Original languageEnglish
Pages (from-to)39
Number of pages1
JournalGastroenterology
Volume144
Issue number5
Publication statusPublished - 1 May 2013

Keywords

  • procarbazine
  • colorectal cancer
  • follow up
  • Netherlands
  • cancer risk
  • gastrointestinal disease
  • patient
  • human
  • risk
  • Hodgkin disease
  • survivor
  • regression analysis
  • transverse colon
  • neoplasm
  • drug megadose
  • colon cancer
  • confidence interval
  • standardized incidence ratio
  • death
  • population
  • rectum
  • hazard ratio
  • lung cancer
  • lymph node
  • rectum cancer
  • breast
  • second cancer
  • spleen
  • radiotherapy
  • high risk population
  • chemotherapy

Fingerprint

Dive into the research topics of 'Subsite specific risk of colorectal cancer risk in patients treated for hodgkin's lymphoma: A long-term follow-up study in the Netherlands'. Together they form a unique fingerprint.

Cite this