Mycoplasma Pulmonis and Lymphoma in a Methanol Bioassay

Pursuant to our commentaries concerning lymphomas and Mycoplasma pulmonis in bioassays of aspartame, methanol, and methyl tert-butyl ether in rats, ^15,16 we direct interested readers to a partial review of the methanol study ¹⁷ in which 3 National Toxicology Program (NTP) pathologists examined selected slides from 100 male rats in each of 0 ppm and 20,000 ppm dose groups. The review team’s report ⁹ states, “The histopathologic diagnoses by the NTP pathologists for a number of lesions… differed from those of the [study pathologist].” The NTP pathologists “occasionally diagnosed ‘leukemia’ or ‘lymphoma’ of the lung, but at a lower frequency than the original findings,” and diagnosed leukemia or lymphoma less frequently in lymph nodes. This report led to the unprecedented and commendable action by the US Environmental Protection Agency to hold “four of its ongoing Integrated Risk Information System assessments pending a review,” including those of methanol and methyl tert-butyl ether, and to undertake “a thorough review of all ongoing and previous chemical assessments to determine which, if any, relied substantially on cancer testing” by the laboratory that conducted these bioassays. ³

The NTP team’s report does not include specific findings that would allow assessment of the possibility that results of a full audit and pathology peer review would alter the conclusion that methanol induced lymphoma, ¹⁷ stating that the “findings are preliminary and intended as a basis for recommendations and are not intended to reach conclusions about any possible or reported effect of the chemical under study.” ⁹ We consider it in the public interest to promote resolution of the scientific questions that have been raised about the methanol and other bioassays. ^{2,4,6 –8,10 –13,15,16} We therefore obtained the results of the slide review from the National Institute of Environmental Health Sciences via a Freedom of Information Act request and analyzed them for differences between the NTP pathologists' diagnoses and those of the study pathologist. We tabulated diagnoses of lymphoma, leukemia, and histiocytic sarcoma in thymus, lung, liver, spleen, and lymph nodes and inflammatory lesions in the nose, lung, and ear. Diagnoses of lymphoma or leukemia were considered to agree without regard to further classification; thus, diagnoses of lymphoimmunoblastic lymphoma and malignant lymphoma were considered to agree, as were diagnoses of myeloid and mononuclear leukemia. Diagnoses of any form of inflammation also were considered to agree. In 11 instances, the NTP pathologists could not compare diagnoses with the study pathologist’s because the listed organ was not present in the section or was too autolyzed; these were considered disagreements. Finally, we compared the NTP pathologists' tabulated diagnoses with their handwritten slide review findings. Where discrepancies existed, we considered the handwritten diagnoses to be correct (2 lungs). Organs for which the tabulated diagnosis could not be verified because the handwritten entry was absent or illegible (1 spleen, 2 thymuses, and 2 lymph nodes) were excluded.

Among 152 hemopoietic neoplasm diagnoses by the study pathologist, the NTP pathologists agreed in 82 cases (54%) and with 38 of 105 lymphoma diagnoses (36%) (Table 1). The NTP pathologists made 9 additional lymphoma diagnoses. Agreement was lowest for lymphoma in the lung; the NTP pathologists agreed with 11 of 52 lung lymphoma diagnoses (21%) for both dose groups. Table 2 shows lymphoma diagnoses by rat. The NTP pathologists agreed with the diagnosis of lymphoma in 15 of 55 rats (27%), and their findings do not indicate a significant difference in lymphoma occurrence between dose groups (Yates corrected χ², P = .612). They identified 1 rat with lymphoma affecting only the lung, whereas 23 such rats were recorded by the study pathologist. The NTP pathologists diagnosed about twice as many inflammatory lesions in the nose, ear, and lung as the study pathologist (Table 3), and from their findings, 100% of the rats in both dose groups had inflammatory lesions in at least one of these sites. Their characterizations of these lesions, such as “chronic, moderate to severe, bronchopneumonia with marked lymphoplasmacytic infiltrates and proliferation (peribronchial and peribronchiolar), bronchiectasis, consolidation, and multifocal suppurative inflammation,” ⁹ are typical of M. pulmonis disease. ^1,5,14,18 The pathologists also noted that “leukemia or lymphoma was sometimes difficult to distinguish from the intense, marked lymphocytic infiltrates related to the chronic inflammation of the lung.” ⁹ These results are consistent with our view that lesions of M. pulmonis disease were interpreted as lymphoma in this bioassay. ^15,16

Table 1.

Methanol Study Hematopoietic Neoplasm Diagnoses by Organ

		Hemopoietic a					Lymphoma b
		Cases		Agreement			Cases		Agreement
Organ	Dose	SP	NTP	No.	%	Additional c	SP3	NTP	No.	%	Additional c
Thymus	0	3	3	3	100	0	1	1	1	100	0
	20 000	1	1	1	100	0	1	1	1	100	0
	All	4	4	4	100	0	2	2	2	100	0
Lung	0	24	10	10	42	0	16	5	4	25	1
	20 000	38	9	8	21	1	36	7	7	19	0
	All	62	19	18	29	1	52	12	11	21	1
Liver	0	14	15	14	100	1	3	5	3	100	2
	20 000	7	8	6	86	2	5	4	3	60	1
	All	21	23	20	95	3	8	9	6	75	3
Spleen	0	13	14	12	92	2	3	5	3	100	2
	20 000	13	9	7	54	2	11	5	4	36	1
	All	26	23	19	73	4	14	11	7	50	3
Lymph node	0	14	12	11	79	1	6	5	3	50	2
	20 000	25	10	10	40	0	23	9	9	39	0
	All	39	22	21	54	1	29	14	12	41	2
Totals	0	68	54	50	74	4	29	21	14	48	7
	20 000	84	37	32	38	5	76	26	24	32	2
	All	152	91	82	54	9	105	47	38	36	9

Doses in parts per million. SP, study pathologist; NTP, National Toxicology Program.

^a Lymphoma (any form), leukemia (any form), and histiocytic sarcoma.

^b Any form.

^c Additional diagnoses by NTP pathologists.

Table 2.

Methanol Study Lymphoma Diagnoses by Rat

		Cases		Agreement
Distribution	Dose	SP	NTP	No.	%	Additional a
Any organ b	0	17	7	5	29	2
	20 000	38	10	10	26	0
	All	55	17	15	27	2
Lung only c	0	10	0	0	0	0
	20 000	13	1	0	0	1
	All	23	1	0	0	1
Lung + other d	0	6	5	3	50	2
	20 000	23	6	6	26	0
	All	29	11	9	31	2

Any form of lymphoma. Doses in parts per million. SP, study pathologist; NTP, National Toxicology Program.

^a Additional cases identified by NTP pathologists.

^b Lymphoma diagnosed in any organ (includes rats with lymphoma not affecting the lung).

^c Lymphoma diagnosed only in the lung.

^d Lymphoma diagnosed in the lung and any other organ.

Table 3.

Methanol Study Inflammatory Lesions by Organ

		Cases
Organ	Dose	SP	NTP
Lung	0	52	93
	20 000	31	88
	All	83	181
Nose	0	37	96
	20 000	24	85
	All	61	181
Ear(s)	0	70	93
	20 000	60	83
	All	130	176
Totals	0	159	282
	20 000	115	256
	All	274	538

Any inflammatory lesion. Doses in parts per million. SP, study pathologist; NTP, National Toxicology Program.

In reviewing the diagnoses, we also noted disagreements regarding other tumors—for example, squamous cell carcinoma in the ear, another tumor reported to be significantly increased by methanol in this bioassay. ¹⁷ Table 4 shows that, as with the findings for lymphoma, the NTP pathologists diagnosed fewer cases of this tumor, and their findings do not indicate a treatment effect (Yates corrected χ², P = .591). The disagreements may be related to squamous epithelial hyperplasia associated with otitis, ⁹ although the report does not specifically state this to be the case.

Table 4.

Methanol Study Diagnoses of Squamous Cell Carcinoma of the Ear

	Cases		Agreement
Dose	SP	NTP	No.	%
0	9	6	6	67
20 000	24	9	9	38
All	33	15	15	45

Doses in parts per million. SP, study pathologist; NTP, National Toxicology Program.