12. Quarterly Financial Data (Unaudited)

The following financial information reflects all normal recurring adjustments, which are, in the opinion of management, necessary for a fair statement of the results of the interim periods. Summarized quarterly data for the years ended December 31, 2017 and 2016 are as follows (in thousands, except per share data):

2017 Quarters	First Quarter			Second Quarter			Third Quarter			Fourth Quarter
	(as revised)
Revenue	$	6,094		$	5,713		$	9,906		$	21,688
Operating expenses	$	69,470		$	25,402		$	26,013		$	42,986
Income (loss) from operations	$	(63,376	)	$	(19,689	)	$	(16,107	)	$	(21,298	)
Net income (loss)	$	(63,856	)	$	(20,359	)	$	(17,637	)	$	(19,707	)
Net (loss) income per preferred share – basic and diluted (1) (2) (3)	$	(2.21	)	$	(0.70	)	$	(0.01	)	$	—
Net loss per common share – basic and diluted (1) (2) (3)		—			—		$	(0.33	)	$	(0.30	)

2016 Quarters	First Quarter			Second Quarter			Third Quarter			Fourth Quarter
Revenue	$	—		$	—		$	—		$	—
Operating expenses	$	16,049		$	13,706		$	19,902		$	33,855
Income (loss) from operations	$	(16,049	)	$	(13,706	)	$	(19,902	)	$	(33,855	)
Net income (loss)	$	(15,962	)	$	(13,615	)	$	(19,809	)	$	(33,831	)
Net loss per preferred share – basic and diluted (1) (3)	$	(0.55	)	$	(0.47	)	$	(0.69	)	$	(1.17	)

_______________________________________________________________________

(1)	We computed net loss per share independently for each of the quarters presented. Therefore, the sum of the quarterly net loss per share will not necessarily equal the total for the year.

(2)

For the purposes of calculating EPS under the two-class method since our IPO in July 2017, we have allocated the net loss between the common stock and the Series A convertible preferred stock for the three-month period ended September 30, 2017. We determined it was appropriate to allocate losses to the Series A convertible preferred stock because it was the lowest form of subordinated equity during such period and because Ionis, the sole holder of the Series A convertible preferred stock, was absorbing our losses during such period. Basic EPS for each class of stock is computed by dividing total distributable losses applicable to preferred and common stock, including the 6% cumulative dividend contractually due to Series A convertible preferred shareholders, by the weighted-average of preferred and common shares outstanding during the requisite period. The cumulative preferred stock dividend was not paid upon completion of the IPO because the IPO was not a liquidation event or a change in control. Prior to the IPO, the 6% cumulative Series A convertible preferred stock dividend was considered as required under the two-class method regardless of whether those dividends were actually distributed.

The following table summarizes the distributable losses for the quarter ended September 30, 2017 (in thousands):

	September 30, 2017
	(as revised)
Net loss	$	(17,637	)
Preferred stock dividend		(1,791	)
Distributable losses	$	(19,428	)

The following table summarizes the reconciliation of weighted-average shares outstanding used in the calculation of basic EPS for the quarter ended September 30, 2017:

Determination of shares:	September 30, 2017
Weighted-average preferred shares outstanding		5,651,323
Weighted-average common shares outstanding		53,522,615
Total weighted-average shares outstanding		59,173,938

The following table summarizes the calculation of basic EPS for the quarter ended September 30, 2017 (in thousands, except per share amounts):

	September 30, 2017
	(as revised)
Losses attributable to preferred shares	$	(1,855	)
Less: Assumed dividend to preferred shares		1,791
Income (losses) allocated to preferred shares	$	(64	)
Weighted-average preferred shares outstanding		5,651,323
Basic income (loss) per preferred share	$	(0.01	)

Losses allocated to common shares	$	(17,573	)
Weighted-average common shares outstanding		53,522,615
Basic loss per common share	$	(0.33	)

(3)	We did not include dilutive common equivalent shares in the computation of diluted net loss per share because the effect would have been antidilutive.