Risk of second primary cancers after a diagnosis of first primary cancer: A pan-cancer analysis and Mendelian randomization study

  1. Xiaohao Ruan
  2. Da Huang
  3. Yongle Zhan
  4. Jingyi Huang
  5. Jinlun Huang
  6. Ada Tsui-Lin Ng
  7. James Hok-Leung Tsu
  8. Rong Na  Is a corresponding author
  1. Department of Urology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, China
  2. Division of Urology, Department of Surgery, Queen Mary Hospital, The University of Hong Kong, China
2 figures, 3 tables and 3 additional files

Figures

Figure 1 with 4 supplements
Study design and work flowchart.
Figure 1—figure supplement 1
Scatter plot.

(A) Oral and pharyngeal cancer on non-Hodgkin lymphoma; (B) oral and pharyngeal cancer on myeloma; (C) esophageal cancer on myeloma; (D) colorectal cancer on liver cancer; (E) pancreatic cancer on gallbladder cancer; (F) pancreatic cancer on small intestin cancer; (G) pancreatic cancer on lary cancer; (H) lung cancer on bone cancer; (I) lung cancer on female breast cancer; (J) lung cancer on Hodgkin lymphoma; (K) melanoma on female breast cancer; (L) female breast cancer on liver cancer; (M) ovary cancer on soft tissue cancer; (N) prostate cancer on colorectal cancer; (O) prostate cancer on pancreatic cancer; (P) prostate cancer on melanoma; (Q) kidney cancer on lung cancer; (R) kidney cancer on melanoma; (S) kidney cancer on non-hodgkin lymphoma; (T) kidney cancer on myeloma; (U) thyroid cancer on liver; (V) thyroid cancer on colorectal cancer; (W) myeloma on lung cancer.

Figure 1—figure supplement 2
Funnel plot.

(A) Oral and pharyngeal cancer on non-Hodgkin lymphoma; (B) oral and pharyngeal cancer on myeloma; (C) esophageal cancer on myeloma; (D) colorectal cancer on liver cancer; (E) pancreatic cancer on gallbladder cancer; (F) pancreatic cancer on small intestin cancer; (G) pancreatic cancer on lary cancer; (H) lung cancer on bone cancer; (I) lung cancer on female breast cancer; (J) lung cancer on Hodgkin lymphoma; (K) melanoma on female breast cancer; (L) female breast cancer on liver cancer; (M) ovary cancer on soft tissue cancer; (N) prostate cancer on colorectal cancer; (O) prostate cancer on pancreatic cancer; (P) prostate cancer on melanoma; (Q) kidney cancer on lung cancer; (R) kidney cancer on melanoma; (S) kidney cancer on non-hodgkin lymphoma; (T) kidney cancer on myeloma; (U) thyroid cancer on liver; (V) thyroid cancer on colorectal cancer; (W) myeloma on lung cancer.

Figure 1—figure supplement 3
Forrest plot.

(A) Oral and pharyngeal cancer on non-Hodgkin lymphoma; (B) oral and pharyngeal cancer on myeloma; (C) esophageal cancer on myeloma; (D) colorectal cancer on liver cancer; (E) pancreatic cancer on gallbladder cancer; (F) pancreatic cancer on small intestin cancer; (G) pancreatic cancer on lary cancer; (H) lung cancer on bone cancer; (I) lung cancer on female breast cancer; (J) lung cancer on Hodgkin lymphoma; (K) melanoma on female breast cancer; (L) female breast cancer on liver cancer; (M) ovary cancer on soft tissue cancer; (N) prostate cancer on colorectal cancer; (O) prostate cancer on pancreatic cancer; (P) prostate cancer on melanoma; (Q) kidney cancer on lung cancer; (R) kidney cancer on melanoma; (S) kidney cancer on non-hodgkin lymphoma; (T) kidney cancer on myeloma; (U) thyroid cancer on liver; (V) thyroid cancer on colorectal cancer; (W) myeloma on lung cancer.

Figure 1—figure supplement 4
Leave-one-out plot.

(A) Oral and pharyngeal cancer on non-Hodgkin lymphoma; (B) oral and pharyngeal cancer on myeloma; (C) esophageal cancer on myeloma; (D) colorectal cancer on liver cancer; (E) pancreatic cancer on gallbladder cancer; (F) pancreatic cancer on small intestin cancer; (G) pancreatic cancer on lary cancer; (H) lung cancer on bone cancer; (I) lung cancer on female breast cancer; (J) lung cancer on Hodgkin lymphoma; (K) melanoma on female breast cancer; (L) female breast cancer on liver cancer; (M) ovary cancer on soft tissue cancer; (N) prostate cancer on colorectal cancer; (O) prostate cancer on pancreatic cancer; (P) prostate cancer on melanoma; (Q) kidney cancer on lung cancer; (R) kidney cancer on melanoma; (S) kidney cancer on non- hodgkin lymphoma; (T) kidney cancer on myeloma; (U) thyroid cancer on liver; (V) thyroid cancer on colorectal cancer; (W) myeloma on lung cancer.

Heat-map of standard incidence ratio for First primary cancers (FPC, horizontal, cancer listed in the bottom) and second primary cancers (SPC, vertical, cancer listed on the right) in the SEER 18 registry (2000–2016).

The standardized incidence observed/expected (O/E) ratios and the corresponding 95% confidence intervals were derived using the “MP-SIR” session of SEER*stat (version 8.3.8).

Tables

Table 1
Summary of the cancer-specific instrument variables used in this study (European ancestry).
Cancer typeGWAS SourcePMIDNumber of SNPs*CasesTotal populationVariance explained, R2, %F-statistics
Oral and pharynxGCST003857277498457603412,6192.22283.68
LarynxGCST01028532276964139448870.2813.72
EsophagusGCST00374027527254510,27927,4380.72198.94
StomachGCST0029902609886612043202,5330.0234.44
Colon and rectumGCST00301726151821818,29937,9550.89340.75
PancreasGCST0054342942260414904021,5364.28962.28
MelanomaGCST00414228212542176628293,1930.29852.68
LungGCST00474828604730729,26685,7160.55474.00
Male BreastGCST011526327856462219068361.2787.91
Female BreastGCST00498829059683104122,977228,9514.9511917.81
Cervix uteriGCST004833288067491286693470.4542.24
OvaryGCST00274825581431918,53069,7451.16818.42
ProstateGCST006085298920167179,148140,2545.478111.66
BladderGCST002240241631277230562068.46572.81
KidneyGCST00471028598434810,78431,1901.11349.99
ThyroidGCST0041442819514263001290,5510.10290.83
MyelomaGCST00448327363682139866249,0540.31774.24
  1. Note: See Supplementary file 2a: the list of SNPs included in the final calculation for each phenotype.

  2. No study found in European: bone and joint, brain, liver, small intestine, gallbladder, renal pelvis and ureter.

  3. Heterogeneity: lymphoma, leukemia.

  4. Too few SNPs: larynx, stomach, male breast cancer.

  5. *

    Number of SNPs included in the final calculation of PRS in our study; not necessarily the total number of SNPs from the source due to the filtering steps discussed in the main text and germline data availability.

  6. MR would be performed based on at least 4 SNPs.

Table 2
Concordant causality between Mendelian randomization results and SEER analysis.
Cancer type(first primary cancer)Cancer type (second primary cancer)
Positive CausalityNegative CausalityNo Causality(statistical power ≥0.80)
Oral and pharynxNon-hodgkin lymphoma-Female breast, Prostate
Esophagus---
Colon and rectum---
PancreasSmall intestin-Melanoma
MelanomaFemale breast-Colon and Rectum
Lung-Female Breast-
Female Breast---
OvarySoft tissue--
Prostate-Colon and RectumNon-Hodgkin lymphoma
Bladder--Female breast, Non-hodgkin lymphoma
KidneyLung and Bronchus, Melanoma, Non-hodgkin lymphoma, Myeloma--
Thyroid---
Myeloma-Lung and Bronchus-
  1. See Supplementary file 2c for unconcordant causality result; Supplementary file 2d-p for the details of Mendelian randomization results (IVW and MR-Egger).

Table 3
Sensitivity and robust analysis of the concordant causality with outlier-filtering approaches.
Exp-outRadial IVWRadial MR-EggerCAUSE
OR (95% CI)pOR (95% CI)pOR (95% CI)p
Oropharynx-NHL1.21 (1.13–1.30)6.00×10–31.21 (1.13–1.30)0.521.17 (1.05–1.31)8.90×10–3
Pancreas-Intestin1.05 (0.83–1.32)0.692.39 (0.91–6.31)0.101.03 (0.79–1.35)1.00
Melanoma-Breast1.08 (1.00–1.15)0.061.05 (0.84–1.32)0.691.04 (0.95–1.16)0.75
Lung-Breast0.86 (0.79–0.93)9.76×10–30.77 (0.54–1.08)0.190.89 (0.76–1.04)0.23
Ovary-Soft1.39 (1.22–1.58)1.07×10–31.3 (0.74–2.29)0.391.36 (1.16–1.58)0.01
Prostate-CRC1.00 (0.95–1.04)0.940.88 (0.80–0.98)0.020.99 (0.93–1.04)0.99
Kidney-Lung1.17 (1.08–1.27)6.60×10–30.94 (0.62–1.43)0.781.16 (1.02–1.31)0.05
Kidney-Melanoma1.33 (1.02–1.73)0.040.56 (0.18–1.80)0.371.25 (0.96–1.73)0.51
Kidney-NHL1.25 (1.11–1.40)7.87×10–31.33 (0.72–2.46)0.391.20 (0.99–1.43)0.09
Kidney-Myeloma1.72 (1.21–2.45)0.020.43 (0.11–1.77)0.291.49 (1.04–2.34)0.02
Myeloma-Lung0.92 (0.86–0.98)0.021.09 (0.79–1.51)0.610.93 (0.86–1.00)0.21
  1. IVW, Inverse variance weighted; CAUSE, Causal Analysis Using Summary Effect estimates; OR, odds ratio; CI, confidence interval; exp, exposure; out, outcome; NHL, non-Hodgkin lymphoma; CRC, colorectal cancer.

Additional files

Supplementary file 1

Standard incidence ratio analysis of different types of cancers in patients with malignancies in SEER database.

https://cdn.elifesciences.org/articles/86379/elife-86379-supp1-v1.xlsx
Supplementary file 2

Mendelian randomization results of FPCs and SPCs in UK Biobank database.

https://cdn.elifesciences.org/articles/86379/elife-86379-supp2-v1.xlsx
MDAR checklist
https://cdn.elifesciences.org/articles/86379/elife-86379-mdarchecklist1-v1.docx

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Xiaohao Ruan
  2. Da Huang
  3. Yongle Zhan
  4. Jingyi Huang
  5. Jinlun Huang
  6. Ada Tsui-Lin Ng
  7. James Hok-Leung Tsu
  8. Rong Na
(2023)
Risk of second primary cancers after a diagnosis of first primary cancer: A pan-cancer analysis and Mendelian randomization study
eLife 12:e86379.
https://doi.org/10.7554/eLife.86379